From: Dmitry Safonov <dima@arista.com>
To: linux-kernel@vger.kernel.org
Cc: Andrei Vagin <avagin@gmail.com>, Dmitry Safonov <dima@arista.com>,
Adrian Reber <adrian@lisas.de>, Andrei Vagin <avagin@openvz.org>,
Andy Lutomirski <luto@kernel.org>,
Andy Tucker <agtucker@google.com>, Arnd Bergmann <arnd@arndb.de>,
Christian Brauner <christian.brauner@ubuntu.com>,
Cyrill Gorcunov <gorcunov@openvz.org>,
Dmitry Safonov <0x7f454c46@gmail.com>,
"Eric W. Biederman" <ebiederm@xmission.com>,
"H. Peter Anvin" <hpa@zytor.com>, Ingo Molnar <mingo@redhat.com>,
Jeff Dike <jdike@addtoit.com>, Oleg Nesterov <oleg@redhat.com>,
Pavel Emelyanov <xemul@virtuozzo.com>,
Shuah Khan <shuah@kernel.org>,
Thomas Gleixner <tglx@linutronix.de>,
containers@lists.linux-foundation.org, criu@openvz.org,
linux-api@vger.kernel.org, x86@kernel.org
Subject: [PATCH 31/32] x86/vdso: Align VDSO functions by CPU L1 cache line
Date: Wed, 6 Feb 2019 00:11:05 +0000 [thread overview]
Message-ID: <20190206001107.16488-32-dima@arista.com> (raw)
In-Reply-To: <20190206001107.16488-1-dima@arista.com>
From: Andrei Vagin <avagin@gmail.com>
After performance testing VDSO patches a noticeable 20% regression was
found on gettime_perf selftest with a cold cache.
As it turns to be, before time namespaces introduction, VDSO functions
were quite aligned to cache lines, but adding a new code to adjust
timens offset inside namespace created a small shift and vdso functions
become unaligned on cache lines.
Add align to vdso functions with gcc option to fix performance drop.
Coping the resulting numbers from cover letter:
Hot CPU cache (more gettime_perf.c cycles - the better):
| before | CONFIG_TIME_NS=n | host | inside timens
--------|------------|------------------|-------------|-------------
cycles | 139887013 | 139453003 | 139899785 | 128792458
diff (%)| 100 | 99.7 | 100 | 92
Cold cache (lesser tsc per gettime_perf_cold.c cycle - the better):
| before | CONFIG_TIME_NS=n | host | inside timens
--------|------------|------------------|-------------|-------------
tsc | 6748 | 6718 | 6862 | 12682
diff (%)| 100 | 99.6 | 101.7 | 188
Measured on Intel(R) Core(TM) i5-6300U CPU @ 2.40GHz
Co-developed-by: Dmitry Safonov <dima@arista.com>
Signed-off-by: Andrei Vagin <avagin@gmail.com>
Signed-off-by: Dmitry Safonov <dima@arista.com>
---
arch/x86/entry/vdso/Makefile | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/x86/entry/vdso/Makefile b/arch/x86/entry/vdso/Makefile
index 4e1659619e7e..2cac4660db05 100644
--- a/arch/x86/entry/vdso/Makefile
+++ b/arch/x86/entry/vdso/Makefile
@@ -4,6 +4,7 @@
#
KBUILD_CFLAGS += $(DISABLE_LTO) -ffunction-sections
+KBUILD_CFLAGS += -falign-functions=$(CONFIG_X86_L1_CACHE_SHIFT)
KASAN_SANITIZE := n
UBSAN_SANITIZE := n
OBJECT_FILES_NON_STANDARD := y
--
2.20.1
next prev parent reply other threads:[~2019-02-06 0:12 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-06 0:10 [PATCH 00/32] kernel: Introduce Time Namespace Dmitry Safonov
2019-02-06 0:10 ` [PATCH 01/32] ns: " Dmitry Safonov
2019-02-06 0:10 ` [PATCH 02/32] timens: Add timens_offsets Dmitry Safonov
2019-02-06 0:10 ` [PATCH 03/32] timens: Introduce CLOCK_MONOTONIC offsets Dmitry Safonov
2019-02-07 21:40 ` Thomas Gleixner
2019-02-08 9:02 ` Andrei Vagin
2019-02-08 9:46 ` Thomas Gleixner
2019-02-06 0:10 ` [PATCH 04/32] timens: Introduce CLOCK_BOOTTIME offset Dmitry Safonov
2019-02-06 0:10 ` [PATCH 05/32] timerfd/timens: Take into account ns clock offsets Dmitry Safonov
2019-02-06 8:52 ` Cyrill Gorcunov
2019-02-06 8:55 ` Cyrill Gorcunov
2019-02-07 6:38 ` Andrei Vagin
2019-02-06 0:10 ` [PATCH 06/32] posix-timers/timens: Take into account " Dmitry Safonov
2019-02-06 0:10 ` [PATCH 07/32] timens/kernel: Take into account timens clock offsets in clock_nanosleep Dmitry Safonov
2019-02-08 7:56 ` Thomas Gleixner
2019-02-06 0:10 ` [PATCH 08/32] timens: Shift /proc/uptime Dmitry Safonov
2019-02-06 0:10 ` [PATCH 09/32] x86/vdso2c: Correct err messages on file opening Dmitry Safonov
2019-02-06 0:10 ` [PATCH 10/32] x86/vdso2c: Convert iterator to unsigned Dmitry Safonov
2019-02-06 0:10 ` [PATCH 11/32] x86/vdso/Makefile: Add vobjs32 Dmitry Safonov
2019-02-06 0:10 ` [PATCH 12/32] x86/vdso/timens: Add offsets page in vvar Dmitry Safonov
2019-02-06 0:10 ` [PATCH 13/32] x86/vdso: Build timens .so(s) Dmitry Safonov
2019-02-06 0:10 ` [PATCH 14/32] x86/VDSO: Build VDSO with -ffunction-sections Dmitry Safonov
2019-02-06 0:10 ` [PATCH 15/32] x86/vdso2c: Optionally produce linker script for vdso entries Dmitry Safonov
2019-02-06 0:10 ` [PATCH 16/32] x86/vdso: Generate vdso{,32}-timens.lds Dmitry Safonov
2019-02-07 8:31 ` Rasmus Villemoes
2019-02-07 16:11 ` Dmitry Safonov
2019-02-08 9:57 ` Thomas Gleixner
2019-02-08 15:18 ` Dmitry Safonov
2019-03-27 18:00 ` Andrei Vagin
2019-03-27 18:06 ` [PATCH RFC] x86/asm: Introduce static_retcall(s) Andrei Vagin
2019-03-27 18:06 ` [PATCH RFC] vdso: introduce timens_static_branch Andrei Vagin
2019-02-06 0:10 ` [PATCH 17/32] x86/vdso2c: Sort vdso entries by addresses for linker script Dmitry Safonov
2019-02-06 0:10 ` [PATCH 18/32] x86/vdso.lds: Align !timens (host's) vdso.so entries Dmitry Safonov
2019-02-06 0:10 ` [PATCH 19/32] x86/vdso2c: Align LOCAL symbols between vdso{-timens,}.so Dmitry Safonov
2019-02-06 0:10 ` [PATCH 20/32] x86/vdso: Initialize timens 64-bit vdso Dmitry Safonov
2019-02-06 0:10 ` [PATCH 21/32] x86/vdso: Switch image on setns()/unshare()/clone() Dmitry Safonov
2019-02-06 0:10 ` [PATCH 22/32] timens: Add align for timens_offsets Dmitry Safonov
2019-02-06 0:10 ` [PATCH 23/32] timens/fs/proc: Introduce /proc/pid/timens_offsets Dmitry Safonov
2019-02-06 0:10 ` [PATCH 24/32] selftest/timens: Add Time Namespace test for supported clocks Dmitry Safonov
2019-02-06 0:10 ` [PATCH 25/32] selftest/timens: Add a test for timerfd Dmitry Safonov
2019-02-06 0:11 ` [PATCH 26/32] selftest/timens: Add a test for clock_nanosleep() Dmitry Safonov
2019-02-06 0:11 ` [PATCH 27/32] selftest/timens: Add procfs selftest Dmitry Safonov
2019-02-06 0:11 ` [PATCH 28/32] selftest/timens: Add timer offsets test Dmitry Safonov
2019-02-06 0:11 ` [PATCH 29/32] selftests: Add a simple perf test for clock_gettime() Dmitry Safonov
2019-02-06 0:11 ` [PATCH 30/32] selftest/timens: Check that a right vdso is mapped after fork and exec Dmitry Safonov
2019-02-06 0:11 ` Dmitry Safonov [this message]
2019-02-06 0:11 ` [PATCH 32/32] x86/vdso: Restrict splitting VVAR VMA Dmitry Safonov
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=20190206001107.16488-32-dima@arista.com \
--to=dima@arista.com \
--cc=0x7f454c46@gmail.com \
--cc=adrian@lisas.de \
--cc=agtucker@google.com \
--cc=arnd@arndb.de \
--cc=avagin@gmail.com \
--cc=avagin@openvz.org \
--cc=christian.brauner@ubuntu.com \
--cc=containers@lists.linux-foundation.org \
--cc=criu@openvz.org \
--cc=ebiederm@xmission.com \
--cc=gorcunov@openvz.org \
--cc=hpa@zytor.com \
--cc=jdike@addtoit.com \
--cc=linux-api@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@redhat.com \
--cc=oleg@redhat.com \
--cc=shuah@kernel.org \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
--cc=xemul@virtuozzo.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;
as well as URLs for NNTP newsgroup(s).