From:   Uros Bizjak <ubizjak@gmail.com>
To:     linux-mm@kvack.org, linux-kernel@vger.kernel.org
Cc:     Uros Bizjak <ubizjak@gmail.com>,
        Andrew Morton <akpm@linux-foundation.org>
Subject: [PATCH] mm/rmap: use atomic_try_cmpxchg in set_tlb_ubc_flush_pending
Date:   Mon, 27 Feb 2023 22:42:28 +0100
Message-Id: <20230227214228.3533299-1-ubizjak@gmail.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Use atomic_try_cmpxchg instead of atomic_cmpxchg (*ptr, old, new) == old in
set_tlb_ubc_flush_pending. 86 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.

No functional change intended.

Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
---
 mm/rmap.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/mm/rmap.c b/mm/rmap.c
index 15ae24585fc4..f964c197fc3c 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -644,7 +644,7 @@ void try_to_unmap_flush_dirty(void)
 static void set_tlb_ubc_flush_pending(struct mm_struct *mm, bool writable)
 {
 	struct tlbflush_unmap_batch *tlb_ubc = &current->tlb_ubc;
-	int batch, nbatch;
+	int batch;
 
 	arch_tlbbatch_add_mm(&tlb_ubc->arch, mm);
 	tlb_ubc->flush_required = true;
@@ -662,11 +662,8 @@ static void set_tlb_ubc_flush_pending(struct mm_struct *mm, bool writable)
 		 * overflow.  Reset `pending' and `flushed' to be 1 and 0 if
 		 * `pending' becomes large.
 		 */
-		nbatch = atomic_cmpxchg(&mm->tlb_flush_batched, batch, 1);
-		if (nbatch != batch) {
-			batch = nbatch;
+		if (!atomic_try_cmpxchg(&mm->tlb_flush_batched, &batch, 1))
 			goto retry;
-		}
 	} else {
 		atomic_inc(&mm->tlb_flush_batched);
 	}
-- 
2.39.2