diff mbox series

[net-next] tcp: Set rcv zerocopy hint correctly if skb last frag is < PAGE_SIZE.

Message ID 20191212225930.233745-1-arjunroy.kdev@gmail.com
State Changes Requested
Delegated to: David Miller
Headers show
Series [net-next] tcp: Set rcv zerocopy hint correctly if skb last frag is < PAGE_SIZE. | expand

Commit Message

Arjun Roy Dec. 12, 2019, 10:59 p.m. UTC 
At present, if the last frag of paged data in a skb has < PAGE_SIZE
data, we compute the recv_skip_hint as being equal to the size of that
frag and the entire next skb.

Instead, just return the runt frag size as the hint.

Signed-off-by: Arjun Roy <arjunroy@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>

---
 net/ipv4/tcp.c | 2 ++
 1 file changed, 2 insertions(+)

Comments

Jakub Kicinski Dec. 15, 2019, 7:22 p.m. UTC | #1 
On Thu, 12 Dec 2019 14:59:30 -0800, Arjun Roy wrote:
> At present, if the last frag of paged data in a skb has < PAGE_SIZE
> data, we compute the recv_skip_hint as being equal to the size of that
> frag and the entire next skb.
> 
> Instead, just return the runt frag size as the hint.

nit: this commit message doesn't really tell us why or what the effects
     are going to be.

> Signed-off-by: Arjun Roy <arjunroy@google.com>
> Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
> Signed-off-by: Eric Dumazet <edumazet@google.com>

WARNING: Missing Signed-off-by: line by nominal patch author 'Arjun Roy <arjunroy.kdev@gmail.com>'

You gotta send it from the same address you signed it off from.
 
> diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
> index 34490d972758..b9623d896469 100644
> --- a/net/ipv4/tcp.c
> +++ b/net/ipv4/tcp.c
> @@ -1904,6 +1904,8 @@ static int tcp_zerocopy_receive(struct sock *sk,
>  	while (length + PAGE_SIZE <= zc->length) {
>  		if (zc->recv_skip_hint < PAGE_SIZE) {
>  			if (skb) {
> +				if (zc->recv_skip_hint > 0)
> +					break;
>  				skb = skb->next;
>  				offset = seq - TCP_SKB_CB(skb)->seq;
>  			} else {
Eric Dumazet Dec. 15, 2019, 7:59 p.m. UTC | #2 
On 12/15/19 11:22 AM, Jakub Kicinski wrote:
> On Thu, 12 Dec 2019 14:59:30 -0800, Arjun Roy wrote:
>> At present, if the last frag of paged data in a skb has < PAGE_SIZE
>> data, we compute the recv_skip_hint as being equal to the size of that
>> frag and the entire next skb.
>>
>> Instead, just return the runt frag size as the hint.
> 
> nit: this commit message doesn't really tell us why or what the effects
>      are going to be.

Right, I added a bit of explanation.

Basically we would hint the application to copy more bytes than needed.

Since zerocopy is best effort, this can target net-next

> 
>> Signed-off-by: Arjun Roy <arjunroy@google.com>
>> Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
>> Signed-off-by: Eric Dumazet <edumazet@google.com>
> 
> WARNING: Missing Signed-off-by: line by nominal patch author 'Arjun Roy <arjunroy.kdev@gmail.com>'
> 
> You gotta send it from the same address you signed it off from.

I sent a v2 on behalf of Arjun.

Thanks.
diff mbox series

Patch

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 34490d972758..b9623d896469 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1904,6 +1904,8 @@  static int tcp_zerocopy_receive(struct sock *sk,
 	while (length + PAGE_SIZE <= zc->length) {
 		if (zc->recv_skip_hint < PAGE_SIZE) {
 			if (skb) {
+				if (zc->recv_skip_hint > 0)
+					break;
 				skb = skb->next;
 				offset = seq - TCP_SKB_CB(skb)->seq;
 			} else {