Re: [PATCH net-next] tcp: properly reset skb->truesize for tx recycling

2019-04-19 Thread David Miller
From: Eric Dumazet Date: Fri, 19 Apr 2019 16:02:03 -0700 > tcp sendmsg() and sendpage() normally advance skb->data_len > and skb->truesize by the payload added to an skb. > > But sendmsg(fd, ..., MSG_ZEROCOPY) has to account for whole pages, > even if a single byte of payload is used in the page

Re: [PATCH net-next] tcp: properly reset skb->truesize for tx recycling

2019-04-19 Thread Soheil Hassas Yeganeh
On Fri, Apr 19, 2019 at 7:02 PM Eric Dumazet wrote: > > tcp sendmsg() and sendpage() normally advance skb->data_len > and skb->truesize by the payload added to an skb. > > But sendmsg(fd, ..., MSG_ZEROCOPY) has to account for whole pages, > even if a single byte of payload is used in the page. > >

[PATCH net-next] tcp: properly reset skb->truesize for tx recycling

2019-04-19 Thread Eric Dumazet
tcp sendmsg() and sendpage() normally advance skb->data_len and skb->truesize by the payload added to an skb. But sendmsg(fd, ..., MSG_ZEROCOPY) has to account for whole pages, even if a single byte of payload is used in the page. This means that we can not assume skb->truesize can be adjusted by