linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Nick Piggin <npiggin@suse.de>
To: Linus Torvalds <torvalds@osdl.org>, Andrew Morton <akpm@osdl.org>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: [patch 2/2] powerpc: native atomic_add_unless
Date: Wed, 18 Jan 2006 07:39:21 +0100	[thread overview]
Message-ID: <20060118063921.GB14608@wotan.suse.de> (raw)
In-Reply-To: <20060118063636.GA14608@wotan.suse.de>

I didn't convert LL/SC architectures so as to "encourage" them
to do atomic_add_unless natively. Here is my probably-wrong attempt
for powerpc.

Should I bring this up on the arch list? (IIRC cross posting 
between there and lkml is discouraged)

Index: linux-2.6/include/asm-powerpc/atomic.h
===================================================================
--- linux-2.6.orig/include/asm-powerpc/atomic.h
+++ linux-2.6/include/asm-powerpc/atomic.h
@@ -8,6 +8,7 @@
 typedef struct { volatile int counter; } atomic_t;
 
 #ifdef __KERNEL__
+#include <linux/compiler.h>
 #include <asm/synch.h>
 #include <asm/asm-compat.h>
 
@@ -176,20 +177,30 @@ static __inline__ int atomic_dec_return(
  * Atomically adds @a to @v, so long as it was not @u.
  * Returns non-zero if @v was not @u, and zero otherwise.
  */
-#define atomic_add_unless(v, a, u)			\
-({							\
-	int c, old;					\
-	c = atomic_read(v);				\
-	for (;;) {					\
-		if (unlikely(c == (u)))			\
-			break;				\
-		old = atomic_cmpxchg((v), c, c + (a));	\
-		if (likely(old == c))			\
-			break;				\
-		c = old;				\
-	}						\
-	c != (u);					\
-})
+static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
+{
+	int t;
+	int dummy;
+
+	__asm__ __volatile__ (
+	LWSYNC_ON_SMP
+"1:	lwarx	%0,0,%2		# atomic_add_unless\n\
+	cmpw	0,%0,%4 \n\
+	beq-	2f \n\
+	add	%1,%3,%0 \n"
+	PPC405_ERR77(0,%2)
+"	stwcx.	%1,0,%2 \n\
+	bne-	1b \n"
+	ISYNC_ON_SMP
+	"\n\
+2:"
+	: "=&r" (t), "=&r" (dummy)
+	: "r" (&v->counter), "r" (a), "r" (u)
+	: "cc", "memory");
+
+	return likely(t != u);
+}
+
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
 
 #define atomic_sub_and_test(a, v)	(atomic_sub_return((a), (v)) == 0)

  reply	other threads:[~2006-01-18  6:39 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-01-18  6:36 [patch 1/2] atomic_add_unless sadness Nick Piggin
2006-01-18  6:39 ` Nick Piggin [this message]
2006-01-18 17:11   ` [patch 2/2] powerpc: native atomic_add_unless Joel Schopp
2006-01-18 17:28     ` Nick Piggin
2006-01-18 21:05       ` Joel Schopp
2006-01-19 14:04         ` Nick Piggin
2006-01-18 16:48 ` [patch 1/2] atomic_add_unless sadness Linus Torvalds
2006-01-18 17:10   ` Nick Piggin

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=20060118063921.GB14608@wotan.suse.de \
    --to=npiggin@suse.de \
    --cc=akpm@osdl.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@osdl.org \
    /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).