netdev - [PATCH net-next 6/7] skbuff: only use pp_magic identifier for a skb' head page

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 22 Sep 2021 17:41:30 +0800
From:   Yunsheng Lin <linyunsheng@...wei.com>
To:     <davem@...emloft.net>, <kuba@...nel.org>
CC:     <netdev@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <linuxarm@...neuler.org>, <hawk@...nel.org>,
        <ilias.apalodimas@...aro.org>, <jonathan.lemon@...il.com>,
        <alobakin@...me>, <willemb@...gle.com>, <cong.wang@...edance.com>,
        <pabeni@...hat.com>, <haokexin@...il.com>, <nogikh@...gle.com>,
        <elver@...gle.com>, <memxor@...il.com>, <edumazet@...gle.com>,
        <alexander.duyck@...il.com>, <dsahern@...il.com>
Subject: [PATCH net-next 6/7] skbuff: only use pp_magic identifier for a skb' head page

As pp_magic is used to identify pp page for a skb's frag page
in previous patch, so do the similar handling for the head
page of a skb too.

And it seems head to frag converting for a skb during GRO and
GSO processing does not need handling when using pp_magic to
identify pp page for a skb' head page and frag page, see
NAPI_GRO_FREE_STOLEN_HEAD for GRO in skb_gro_receive() and
skb_head_frag_to_page_desc() for GSO in skb_segment().

As pp_magic only exist in the head page of a compound page,
and the freeing of a head page for a skb is eventually operated
on the head page of a compound page for both pp and non-pp
page, so use virt_to_head_page() and __page_frag_cache_drain()
in skb_free_head() to avoid unnecessary virt_to_head_page()
calling in page_frag_free().

Signed-off-by: Yunsheng Lin <linyunsheng@...wei.com>
---
 include/linux/skbuff.h | 15 ---------------
 net/core/skbuff.c      | 11 +++++++++--
 2 files changed, 9 insertions(+), 17 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index a2d3b6fe0c32..b77ee060b64d 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -4728,20 +4728,5 @@ static inline void skb_mark_for_recycle(struct sk_buff *skb)
 }
 #endif
 
-static inline bool skb_pp_recycle(struct sk_buff *skb, void *data)
-{
-	struct page *page;
-
-	if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle)
-		return false;
-
-	page = virt_to_head_page(data);
-	if (!page_pool_is_pp_page(page))
-		return false;
-
-	page_pool_return_skb_page(page);
-	return true;
-}
-
 #endif	/* __KERNEL__ */
 #endif	/* _LINUX_SKBUFF_H */
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index db8af3eff255..3718898da499 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -647,9 +647,16 @@ static void skb_free_head(struct sk_buff *skb)
 	unsigned char *head = skb->head;
 
 	if (skb->head_frag) {
-		if (skb_pp_recycle(skb, head))
+		struct page *page = virt_to_head_page(head);
+
+#ifdef CONFIG_PAGE_POOL
+		if (page_pool_is_pp_page(page)) {
+			page_pool_return_skb_page(page);
 			return;
-		skb_free_frag(head);
+		}
+#endif
+
+		__page_frag_cache_drain(page, 1);
 	} else {
 		kfree(head);
 	}
-- 
2.33.0