All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: mm-commits@vger.kernel.org, ubizjak@gmail.com, akpm@linux-foundation.org
Subject: [merged mm-nonmm-stable] iversion-use-atomic64_try_cmpxchg.patch removed from -mm tree
Date: Sun, 11 Sep 2022 21:56:36 -0700	[thread overview]
Message-ID: <20220912045637.78804C433D7@smtp.kernel.org> (raw)


The quilt patch titled
     Subject: iversion: use atomic64_try_cmpxchg)
has been removed from the -mm tree.  Its filename was
     iversion-use-atomic64_try_cmpxchg.patch

This patch was dropped because it was merged into the mm-nonmm-stable branch
of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

------------------------------------------------------
From: Uros Bizjak <ubizjak@gmail.com>
Subject: iversion: use atomic64_try_cmpxchg)
Date: Sun, 21 Aug 2022 21:30:11 +0200

Use atomic64_try_cmpxchg instead of
atomic64_cmpxchg (*ptr, old, new) == old in inode_set_max_iversion_raw,
inode_maybe_inc_version and inode_query_iversion. x86 CMPXCHG instruction
returns success in ZF flag, so this change saves a compare after cmpxchg
(and related move instruction in front of cmpxchg).

Also, try_cmpxchg implicitly assigns old *ptr value to "old" when cmpxchg
fails, enabling further code simplifications.

The loop in inode_maybe_inc_iversion improves from:

    5563:	48 89 ca             	mov    %rcx,%rdx
    5566:	48 89 c8             	mov    %rcx,%rax
    5569:	48 83 e2 fe          	and    $0xfffffffffffffffe,%rdx
    556d:	48 83 c2 02          	add    $0x2,%rdx
    5571:	f0 48 0f b1 16       	lock cmpxchg %rdx,(%rsi)
    5576:	48 39 c1             	cmp    %rax,%rcx
    5579:	0f 84 85 fc ff ff    	je     5204 <...>
    557f:	48 89 c1             	mov    %rax,%rcx
    5582:	eb df                	jmp    5563 <...>

to:

    5563:	48 89 c2             	mov    %rax,%rdx
    5566:	48 83 e2 fe          	and    $0xfffffffffffffffe,%rdx
    556a:	48 83 c2 02          	add    $0x2,%rdx
    556e:	f0 48 0f b1 11       	lock cmpxchg %rdx,(%rcx)
    5573:	0f 84 8b fc ff ff    	je     5204 <...>
    5579:	eb e8                	jmp    5563 <...>

Link: https://lkml.kernel.org/r/20220821193011.88208-1-ubizjak@gmail.com
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 include/linux/iversion.h |   32 +++++++++-----------------------
 1 file changed, 9 insertions(+), 23 deletions(-)

--- a/include/linux/iversion.h~iversion-use-atomic64_try_cmpxchg
+++ a/include/linux/iversion.h
@@ -123,17 +123,12 @@ inode_peek_iversion_raw(const struct ino
 static inline void
 inode_set_max_iversion_raw(struct inode *inode, u64 val)
 {
-	u64 cur, old;
+	u64 cur = inode_peek_iversion_raw(inode);
 
-	cur = inode_peek_iversion_raw(inode);
-	for (;;) {
+	do {
 		if (cur > val)
 			break;
-		old = atomic64_cmpxchg(&inode->i_version, cur, val);
-		if (likely(old == cur))
-			break;
-		cur = old;
-	}
+	} while (!atomic64_try_cmpxchg(&inode->i_version, &cur, val));
 }
 
 /**
@@ -197,7 +192,7 @@ inode_set_iversion_queried(struct inode
 static inline bool
 inode_maybe_inc_iversion(struct inode *inode, bool force)
 {
-	u64 cur, old, new;
+	u64 cur, new;
 
 	/*
 	 * The i_version field is not strictly ordered with any other inode
@@ -211,19 +206,14 @@ inode_maybe_inc_iversion(struct inode *i
 	 */
 	smp_mb();
 	cur = inode_peek_iversion_raw(inode);
-	for (;;) {
+	do {
 		/* If flag is clear then we needn't do anything */
 		if (!force && !(cur & I_VERSION_QUERIED))
 			return false;
 
 		/* Since lowest bit is flag, add 2 to avoid it */
 		new = (cur & ~I_VERSION_QUERIED) + I_VERSION_INCREMENT;
-
-		old = atomic64_cmpxchg(&inode->i_version, cur, new);
-		if (likely(old == cur))
-			break;
-		cur = old;
-	}
+	} while (!atomic64_try_cmpxchg(&inode->i_version, &cur, new));
 	return true;
 }
 
@@ -304,10 +294,10 @@ inode_peek_iversion(const struct inode *
 static inline u64
 inode_query_iversion(struct inode *inode)
 {
-	u64 cur, old, new;
+	u64 cur, new;
 
 	cur = inode_peek_iversion_raw(inode);
-	for (;;) {
+	do {
 		/* If flag is already set, then no need to swap */
 		if (cur & I_VERSION_QUERIED) {
 			/*
@@ -320,11 +310,7 @@ inode_query_iversion(struct inode *inode
 		}
 
 		new = cur | I_VERSION_QUERIED;
-		old = atomic64_cmpxchg(&inode->i_version, cur, new);
-		if (likely(old == cur))
-			break;
-		cur = old;
-	}
+	} while (!atomic64_try_cmpxchg(&inode->i_version, &cur, new));
 	return cur >> I_VERSION_QUERIED_SHIFT;
 }
 
_

Patches currently in -mm which might be from ubizjak@gmail.com are



                 reply	other threads:[~2022-09-12  4:57 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20220912045637.78804C433D7@smtp.kernel.org \
    --to=akpm@linux-foundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mm-commits@vger.kernel.org \
    --cc=ubizjak@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.