From: kernel@xen0n.name
To: linux-mips@vger.kernel.org
Cc: Wang Xuerui <wangxuerui@qiniu.com>,
Huacai Chen <chenhc@lemote.com>,
Jiaxun Yang <jiaxun.yang@flygoat.com>,
James Hogan <jhogan@kernel.org>,
Paul Burton <paul.burton@mips.com>,
Ralf Baechle <ralf@linux-mips.org>,
linux-mips@linux-mips.org
Subject: [RFC PATCH 2/2] MIPS: VDSO: support extcc-based timekeeping
Date: Sun, 24 Feb 2019 17:36:35 +0800 [thread overview]
Message-ID: <20190224093635.1242-3-kernel@xen0n.name> (raw)
In-Reply-To: <20190224093635.1242-1-kernel@xen0n.name>
From: Wang Xuerui <wangxuerui@qiniu.com>
The clocksource bits are ready, just wire things up in VDSO for a
significant user-space timekeeping performance gain. There are several
ABI problems uncovered by vdsotest though, but daily usage of the test
system didn't expose any of them. These inconsistencies will be fixed
in a later commit (presently TODO).
According to vdsotest (formatting is manually added, only the affected
figures are shown):
category before after
-------- ------ -----
clock-gettime-monotonic: syscall: 409 401 nsec/call
clock-gettime-monotonic: libc: 476 141 nsec/call
clock-gettime-monotonic: vdso: 462 123 nsec/call
clock-gettime-realtime: syscall: 405 407 nsec/call
clock-gettime-realtime: libc: 474 142 nsec/call
clock-gettime-realtime: vdso: 457 125 nsec/call
gettimeofday: syscall: 406 407 nsec/call
gettimeofday: libc: 455 121 nsec/call
gettimeofday: vdso: 440 102 nsec/call
The benchmark was run on a single-socket 3A3000 @ 1.4GHz. There is
still plenty of headroom for improvements of course, but let's take
care of the micro-optimizations later.
Signed-off-by: Wang Xuerui <wangxuerui@qiniu.com>
Tested-by: Wang Xuerui <wangxuerui@qiniu.com>
Cc: Huacai Chen <chenhc@lemote.com>
Cc: Jiaxun Yang <jiaxun.yang@flygoat.com>
Cc: James Hogan <jhogan@kernel.org>
Cc: Paul Burton <paul.burton@mips.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: linux-mips@linux-mips.org
---
arch/mips/include/asm/clocksource.h | 1 +
arch/mips/loongson64/common/extcc.c | 3 +--
arch/mips/vdso/gettimeofday.c | 8 ++++++++
3 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/arch/mips/include/asm/clocksource.h b/arch/mips/include/asm/clocksource.h
index 3deb1d0c1a94..6a126a475892 100644
--- a/arch/mips/include/asm/clocksource.h
+++ b/arch/mips/include/asm/clocksource.h
@@ -17,6 +17,7 @@
#define VDSO_CLOCK_NONE 0 /* No suitable clocksource. */
#define VDSO_CLOCK_R4K 1 /* Use the coprocessor 0 count. */
#define VDSO_CLOCK_GIC 2 /* Use the GIC. */
+#define VDSO_CLOCK_EXTCC 3 /* Use the Loongson ExtCC. */
/**
* struct arch_clocksource_data - Architecture-specific clocksource information.
diff --git a/arch/mips/loongson64/common/extcc.c b/arch/mips/loongson64/common/extcc.c
index 702cb389856a..0f6775099411 100644
--- a/arch/mips/loongson64/common/extcc.c
+++ b/arch/mips/loongson64/common/extcc.c
@@ -23,8 +23,7 @@ static struct clocksource extcc_clocksource = {
.read = extcc_read,
.mask = CLOCKSOURCE_MASK(64),
.flags = CLOCK_SOURCE_IS_CONTINUOUS | CLOCK_SOURCE_VALID_FOR_HRES,
- /* TODO later */
- .archdata = { .vdso_clock_mode = VDSO_CLOCK_NONE },
+ .archdata = { .vdso_clock_mode = VDSO_CLOCK_EXTCC },
};
void __init extcc_clocksource_init(void)
diff --git a/arch/mips/vdso/gettimeofday.c b/arch/mips/vdso/gettimeofday.c
index e22b422f282c..92eef8de36a4 100644
--- a/arch/mips/vdso/gettimeofday.c
+++ b/arch/mips/vdso/gettimeofday.c
@@ -17,6 +17,9 @@
#include <asm/io.h>
#include <asm/unistd.h>
#include <asm/vdso.h>
+#ifdef CONFIG_LOONGSON_EXTCC_CLKSRC
+#include <asm/mach-loongson64/extcc.h>
+#endif
#ifdef CONFIG_MIPS_CLOCK_VSYSCALL
@@ -148,6 +151,11 @@ static __always_inline u64 get_ns(const union mips_vdso_data *data)
case VDSO_CLOCK_GIC:
cycle_now = read_gic_count(data);
break;
+#endif
+#ifdef CONFIG_LOONGSON_EXTCC_CLKSRC
+ case VDSO_CLOCK_EXTCC:
+ cycle_now = __extcc_read_ordered();
+ break;
#endif
default:
return 0;
--
2.20.1
next prev parent reply other threads:[~2019-02-24 10:05 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-24 9:36 [RFC PATCH 0/2] MIPS: Loongson: ExtCC clocksource support kernel
2019-02-24 9:36 ` [RFC PATCH 1/2] MIPS: Loongson: add extcc clocksource kernel
2019-02-28 3:12 ` Paul Burton
2019-02-24 9:36 ` kernel [this message]
2019-02-28 3:17 ` [RFC PATCH 2/2] MIPS: VDSO: support extcc-based timekeeping Paul Burton
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=20190224093635.1242-3-kernel@xen0n.name \
--to=kernel@xen0n.name \
--cc=chenhc@lemote.com \
--cc=jhogan@kernel.org \
--cc=jiaxun.yang@flygoat.com \
--cc=linux-mips@linux-mips.org \
--cc=linux-mips@vger.kernel.org \
--cc=paul.burton@mips.com \
--cc=ralf@linux-mips.org \
--cc=wangxuerui@qiniu.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).