[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <tip-824975ef190e7dcb77718d1cc2cb53769b16d918@git.kernel.org>
Date: Fri, 3 Jul 2009 12:43:29 GMT
From: tip-bot for Ingo Molnar <mingo@...e.hu>
To: linux-tip-commits@...r.kernel.org
Cc: linux-kernel@...r.kernel.org, acme@...hat.com, paulus@...ba.org,
hpa@...or.com, mingo@...hat.com, eric.dumazet@...il.com,
a.p.zijlstra@...llo.nl, torvalds@...ux-foundation.org,
efault@....de, arnd@...db.de, dhowells@...hat.com,
fweisbec@...il.com, akpm@...ux-foundation.org, tglx@...utronix.de,
mingo@...e.hu
Subject: [tip:perfcounters/urgent] x86: atomic64: Improve atomic64_add_return()
Commit-ID: 824975ef190e7dcb77718d1cc2cb53769b16d918
Gitweb: http://git.kernel.org/tip/824975ef190e7dcb77718d1cc2cb53769b16d918
Author: Ingo Molnar <mingo@...e.hu>
AuthorDate: Fri, 3 Jul 2009 12:39:07 +0200
Committer: Ingo Molnar <mingo@...e.hu>
CommitDate: Fri, 3 Jul 2009 13:26:42 +0200
x86: atomic64: Improve atomic64_add_return()
Linus noted (based on Eric Dumazet's numbers) that we would
probably be better off not trying an atomic_read() in
atomic64_add_return() but intead intentionally let the first
cmpxchg8b fail - to get a cache-friendly 'give me ownership
of this cacheline' transaction. That can then be followed
by the real cmpxchg8b which sets the value local to the CPU.
Reported-by: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Eric Dumazet <eric.dumazet@...il.com>
Cc: Peter Zijlstra <a.p.zijlstra@...llo.nl>
Cc: Mike Galbraith <efault@....de>
Cc: Paul Mackerras <paulus@...ba.org>
Cc: Arnaldo Carvalho de Melo <acme@...hat.com>
Cc: Frederic Weisbecker <fweisbec@...il.com>
Cc: David Howells <dhowells@...hat.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>
Cc: Arnd Bergmann <arnd@...db.de>
LKML-Reference: <alpine.LFD.2.01.0907021653030.3210@...alhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@...e.hu>
---
arch/x86/lib/atomic64_32.c | 15 ++++++++++++---
1 files changed, 12 insertions(+), 3 deletions(-)
diff --git a/arch/x86/lib/atomic64_32.c b/arch/x86/lib/atomic64_32.c
index 5fc1e2c..6195962 100644
--- a/arch/x86/lib/atomic64_32.c
+++ b/arch/x86/lib/atomic64_32.c
@@ -76,13 +76,22 @@ u64 atomic64_read(atomic64_t *ptr)
*/
u64 atomic64_add_return(u64 delta, atomic64_t *ptr)
{
- u64 old_val, new_val;
+ /*
+ * Try first with a (probably incorrect) assumption about
+ * what we have there. We'll do two loops most likely,
+ * but we'll get an ownership MESI transaction straight away
+ * instead of a read transaction followed by a
+ * flush-for-ownership transaction:
+ */
+ u64 old_val, new_val, real_val = 1ULL << 32;
do {
- old_val = atomic_read(ptr);
+ old_val = real_val;
new_val = old_val + delta;
- } while (atomic64_cmpxchg(ptr, old_val, new_val) != old_val);
+ real_val = atomic64_cmpxchg(ptr, old_val, new_val);
+
+ } while (real_val != old_val);
return new_val;
}
--
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/
Powered by blists - more mailing lists