skb_pp_recycle() is only used by skb_free_head() in skbuff.c, so move it to skbuff.c.
Signed-off-by: Yunsheng Lin linyunsheng@huawei.com --- include/linux/skbuff.h | 7 ------- net/core/skbuff.c | 7 +++++++ 2 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 9fcf534f2d92..28a7b5fbc7b7 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -5048,12 +5048,5 @@ static inline void skb_mark_for_recycle(struct sk_buff *skb) } #endif
-static inline bool skb_pp_recycle(struct sk_buff *skb, void *data) -{ - if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle) - return false; - return page_pool_return_skb_page(virt_to_page(data)); -} - #endif /* __KERNEL__ */ #endif /* _LINUX_SKBUFF_H */ diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 1d9719e72f9d..9b3b19816d2d 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -748,6 +748,13 @@ static void skb_clone_fraglist(struct sk_buff *skb) skb_get(list); }
+static bool skb_pp_recycle(struct sk_buff *skb, void *data) +{ + if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle) + return false; + return page_pool_return_skb_page(virt_to_page(data)); +} + static void skb_free_head(struct sk_buff *skb) { unsigned char *head = skb->head;
Hi Yungsheng
On Fri, 21 Oct 2022 at 05:58, Yunsheng Lin linyunsheng@huawei.com wrote:
skb_pp_recycle() is only used by skb_free_head() in skbuff.c, so move it to skbuff.c.
Signed-off-by: Yunsheng Lin linyunsheng@huawei.com
include/linux/skbuff.h | 7 ------- net/core/skbuff.c | 7 +++++++ 2 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 9fcf534f2d92..28a7b5fbc7b7 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -5048,12 +5048,5 @@ static inline void skb_mark_for_recycle(struct sk_buff *skb) } #endif
-static inline bool skb_pp_recycle(struct sk_buff *skb, void *data) -{
if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle)return false;return page_pool_return_skb_page(virt_to_page(data));-}
#endif /* __KERNEL__ */ #endif /* _LINUX_SKBUFF_H */ diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 1d9719e72f9d..9b3b19816d2d 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -748,6 +748,13 @@ static void skb_clone_fraglist(struct sk_buff *skb) skb_get(list); }
+static bool skb_pp_recycle(struct sk_buff *skb, void *data) +{
if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle)return false;return page_pool_return_skb_page(virt_to_page(data));+}
Any particular reason you are removing the inline hint here? Doing it like this will add an extra function call for every packet (assuming the compiler decided to inline the previous version)
Thanks /Ilias
static void skb_free_head(struct sk_buff *skb) { unsigned char *head = skb->head; -- 2.33.0
On Fri, 21 Oct 2022 09:02:36 +0300 Ilias Apalodimas wrote:
+static bool skb_pp_recycle(struct sk_buff *skb, void *data) +{
if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle)return false;return page_pool_return_skb_page(virt_to_page(data));+}
Any particular reason you are removing the inline hint here?
It's recommended in networking to avoid using the inline keyword unless someone actually checked the compiler output and found the compiler is being stupid. I don't know the full history of this recommendation tho.
Doing it like this will add an extra function call for every packet (assuming the compiler decided to inline the previous version)
Should be fine, tiny static function with one caller, I'd bet it's always inlined, even with -Os.
Hi Jakub On Fri, Oct 21, 2022 at 08:51:38AM -0700, Jakub Kicinski wrote:
On Fri, 21 Oct 2022 09:02:36 +0300 Ilias Apalodimas wrote:
+static bool skb_pp_recycle(struct sk_buff *skb, void *data) +{
if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle)return false;return page_pool_return_skb_page(virt_to_page(data));+}
Any particular reason you are removing the inline hint here?
It's recommended in networking to avoid using the inline keyword unless someone actually checked the compiler output and found the compiler is being stupid. I don't know the full history of this recommendation tho.
Ah thanks, didn't know that. IIRC there was no particular reason. Probably because the majority of the functions in that header are static inlines.
Doing it like this will add an extra function call for every packet (assuming the compiler decided to inline the previous version)
Should be fine, tiny static function with one caller, I'd bet it's always inlined, even with -Os.
Well it's compilers -- I wouldn't bet, but fair enough
Acked-by: Ilias Apalodimas ilias.apalodimas@linaro.org
Thanks /Ilias
Hello:
This patch was applied to netdev/net-next.git (master) by David S. Miller davem@davemloft.net:
On Fri, 21 Oct 2022 10:58:22 +0800 you wrote:
skb_pp_recycle() is only used by skb_free_head() in skbuff.c, so move it to skbuff.c.
Signed-off-by: Yunsheng Lin linyunsheng@huawei.com
include/linux/skbuff.h | 7 ------- net/core/skbuff.c | 7 +++++++ 2 files changed, 7 insertions(+), 7 deletions(-)
Here is the summary with links: - [net-next] net: skb: move skb_pp_recycle() to skbuff.c https://git.kernel.org/netdev/net-next/c/4727bab4e9bb
You are awesome, thank you!
-
Ilias Apalodimas -
Jakub Kicinski -
patchwork-bot+netdevbpf＠kernel.org -
Yunsheng Lin