x86: atomic64: Improve atomic64_xchg()

Remove the read-first logic from atomic64_xchg() and simplify
the loop.

This function was the last user of __atomic64_read() - remove it.

Also, change the 'real_val' assumption from the somewhat quirky
1ULL << 32 value to the (just as arbitrary, but simpler) value
of 0.

Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
LKML-Reference: <tip-05118ab8859492ac9ddda0154cf90e37b0a4a0b0@git.kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

1 files changed, 15 insertions, 6 deletions

diff --git a/arch/x86/lib/atomic64_32.c b/arch/x86/lib/atomic64_32.c
index 6722a092e40..a804f96e90e 100644
--- a/arch/x86/lib/atomic64_32.c
+++ b/arch/x86/lib/atomic64_32.c
@@ -33,14 +33,23 @@ EXPORT_SYMBOL(atomic64_cmpxchg);
  * Atomically xchgs the value of @ptr to @new_val and returns
  * the old value.
  */
-
 u64 atomic64_xchg(atomic64_t *ptr, u64 new_val)
 {
-	u64 old_val;
+	/*
+	 * Try first with a (possibly 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, real_val = 0;
 
 	do {
-		old_val = __atomic64_read(ptr);
-	} while (atomic64_cmpxchg(ptr, old_val, new_val) != old_val);
+		old_val = real_val;
+
+		real_val = atomic64_cmpxchg(ptr, old_val, new_val);
+
+	} while (real_val != old_val);
 
 	return old_val;
 }
@@ -91,13 +100,13 @@ EXPORT_SYMBOL(atomic64_read);
 noinline u64 atomic64_add_return(u64 delta, atomic64_t *ptr)
 {
 	/*
-	 * Try first with a (probably incorrect) assumption about
+	 * Try first with a (possibly 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;
+	u64 old_val, new_val, real_val = 0;
 
 	do {
 		old_val = real_val;