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>, 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>,
Vincenzo Frascino <vincenzo.frascino@arm.com>,
containers@lists.linux-foundation.org, criu@openvz.org,
linux-api@vger.kernel.org, x86@kernel.org
Subject: [PATCHv3 25/27] x86/vdso: Align VDSO functions by CPU L1 cache line
Date: Thu, 25 Apr 2019 17:14:14 +0100 [thread overview]
Message-ID: <20190425161416.26600-26-dima@arista.com> (raw)
In-Reply-To: <20190425161416.26600-1-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 bc0bdbf49397..9b45aafdfcfe 100644
--- a/arch/x86/entry/vdso/Makefile
+++ b/arch/x86/entry/vdso/Makefile
@@ -4,6 +4,7 @@
#
KBUILD_CFLAGS += $(DISABLE_LTO)
+KBUILD_CFLAGS += -falign-functions=$(CONFIG_X86_L1_CACHE_SHIFT)
KASAN_SANITIZE := n
UBSAN_SANITIZE := n
OBJECT_FILES_NON_STANDARD := y
--
2.21.0
next prev parent reply other threads:[~2019-04-25 16:15 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-25 16:13 [PATCHv3 00/27] kernel: Introduce Time Namespace Dmitry Safonov
2019-04-25 16:13 ` [PATCHv3 01/27] ns: " Dmitry Safonov
2019-04-25 19:10 ` Jann Horn
2019-04-25 16:13 ` [PATCHv3 02/27] timens: Add timens_offsets Dmitry Safonov
2019-04-25 16:13 ` [PATCHv3 03/27] timens: Introduce CLOCK_MONOTONIC offsets Dmitry Safonov
2019-04-25 19:52 ` Thomas Gleixner
2019-04-25 16:13 ` [PATCHv3 04/27] timens: Introduce CLOCK_BOOTTIME offset Dmitry Safonov
2019-04-25 20:08 ` Cyrill Gorcunov
2019-04-25 16:13 ` [PATCHv3 05/27] timerfd/timens: Take into account ns clock offsets Dmitry Safonov
2019-04-25 21:28 ` Thomas Gleixner
2019-05-03 7:00 ` Andrei Vagin
2019-04-25 16:13 ` [PATCHv3 06/27] posix-timers/timens: Take into account " Dmitry Safonov
2019-04-25 21:45 ` Thomas Gleixner
2019-04-25 16:13 ` [PATCHv3 07/27] timens/kernel: Take into account timens clock offsets in clock_nanosleep Dmitry Safonov
2019-04-25 16:13 ` [PATCHv3 08/27] timens: Shift /proc/uptime Dmitry Safonov
2019-04-25 16:13 ` [PATCHv3 09/27] x86/vdso2c: Correct err messages on file opening Dmitry Safonov
2019-04-25 16:13 ` [PATCHv3 10/27] x86/vdso2c: Convert iterator to unsigned Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 11/27] x86/vdso/Makefile: Add vobjs32 Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 12/27] x86/vdso: Restrict splitting VVAR VMA Dmitry Safonov
2019-04-25 18:41 ` Jann Horn
2019-04-25 18:46 ` Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 13/27] x86/vdso: Rename vdso_image {.data=>.text} Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 14/27] x86/vdso: Add offsets page in vvar Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 15/27] x86/vdso: Allocate timens vdso Dmitry Safonov
2019-04-25 18:32 ` Jann Horn
2019-04-25 19:05 ` Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 16/27] x86/vdso: Switch image on setns()/unshare()/clone() Dmitry Safonov
2019-04-25 17:53 ` Jann Horn
2019-04-25 18:02 ` Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 17/27] vdso: introduce timens_static_branch Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 18/27] timens: Add align for timens_offsets Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 19/27] timens/fs/proc: Introduce /proc/pid/timens_offsets Dmitry Safonov
2019-04-25 18:16 ` Jann Horn
2019-05-02 6:08 ` Andrei Vagin
2019-04-25 16:14 ` [PATCHv3 20/27] selftest/timens: Add Time Namespace test for supported clocks Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 21/27] selftest/timens: Add a test for timerfd Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 22/27] selftest/timens: Add a test for clock_nanosleep() Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 23/27] selftest/timens: Add procfs selftest Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 24/27] selftest/timens: Add timer offsets test Dmitry Safonov
2019-04-25 16:14 ` Dmitry Safonov [this message]
2019-04-25 16:14 ` [PATCHv3 26/27] selftests: Add a simple perf test for clock_gettime() Dmitry Safonov
2019-04-25 16:14 ` [PATCHv3 27/27] selftest/timens: Check that a right vdso is mapped after fork and exec 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=20190425161416.26600-26-dima@arista.com \
--to=dima@arista.com \
--cc=0x7f454c46@gmail.com \
--cc=adrian@lisas.de \
--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=vincenzo.frascino@arm.com \
--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).