From:   gouhao@uniontech.com
To:     cl@linux.com, penberg@kernel.org, rientjes@google.com,
        iamjoonsoo.kim@lge.com, akpm@linux-foundation.org, vbabka@suse.cz,
        roman.gushchin@linux.dev, 42.hyeyoo@gmail.com,
        keescook@chromium.org
Cc:     linux-mm@kvack.org, linux-kernel@vger.kernel.org,
        linux-hardening@vger.kernel.org
Subject: [PATCH] mm/slub: reduce the calculation times of 'MAX_OBJS_PER_PAGE'
Date:   Thu, 16 Mar 2023 09:25:17 +0800
Message-Id: <20230316012517.10479-1-gouhao@uniontech.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Feedback-ID: bizesmtp:uniontech.com:qybglogicsvr:qybglogicsvr7
Precedence: bulk

From: Gou Hao <gouhao@uniontech.com>

when calling calc_slab_order(), 'slub_min_order'
and 'size' are fixed values, if the condition of
'MAX_OBJS_PER_PAGE' is true, it will be returned from
here every time.

So we can calculate the condition of 'MAX_OBJS_PER_PAGE'
before calling calculate_order().

Signed-off-by: Gou Hao <gouhao@uniontech.com>
---
 mm/slub.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/mm/slub.c b/mm/slub.c
index 39327e98fce3..ed6d797a5cd8 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -4094,9 +4094,6 @@ static inline unsigned int calc_slab_order(unsigned int size,
 	unsigned int min_order = slub_min_order;
 	unsigned int order;
 
-	if (order_objects(min_order, size) > MAX_OBJS_PER_PAGE)
-		return get_order(size * MAX_OBJS_PER_PAGE) - 1;
-
 	for (order = max(min_order, (unsigned int)get_order(min_objects * size));
 			order <= max_order; order++) {
 
@@ -4457,9 +4454,14 @@ static int calculate_sizes(struct kmem_cache *s)
 	size = ALIGN(size, s->align);
 	s->size = size;
 	s->reciprocal_size = reciprocal_value(size);
-	order = calculate_order(size);
 
-	if ((int)order < 0)
+
+	if (order_objects(slub_min_order, size) > MAX_OBJS_PER_PAGE)
+		order = get_order(size * MAX_OBJS_PER_PAGE) - 1;
+	else
+		order = calculate_order(size);
+
+	if ((int)order < 0 || order > slub_max_order)
 		return 0;
 
 	s->allocflags = 0;
-- 
2.34.1