public inbox for [email protected]
 help / color / mirror / Atom feed
From: Willem de Bruijn <[email protected]>
To: Pavel Begunkov <[email protected]>,
	 [email protected],  [email protected]
Cc: Jens Axboe <[email protected]>,
	 [email protected],
	 "David S . Miller" <[email protected]>,
	 Jakub Kicinski <[email protected]>,
	 David Ahern <[email protected]>,
	 Eric Dumazet <[email protected]>,
	 Willem de Bruijn <[email protected]>
Subject: Re: [PATCH net-next 3/5] net: batch zerocopy_fill_skb_from_iter accounting
Date: Fri, 28 Jun 2024 13:06:11 -0400	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <a916f99aa91bc9066411015835cadd5677a454fb.1719190216.git.asml.silence@gmail.com>

Pavel Begunkov wrote:
> Instead of accounting every page range against the socket separately, do
> it in batch based on the change in skb->truesize. It's also moved into
> __zerocopy_sg_from_iter(), so that zerocopy_fill_skb_from_iter() is
> simpler and responsible for setting frags but not the accounting.
> 
> Signed-off-by: Pavel Begunkov <[email protected]>

Reviewed-by: Willem de Bruijn <[email protected]>

> ---
>  net/core/datagram.c | 31 ++++++++++++++++++-------------
>  1 file changed, 18 insertions(+), 13 deletions(-)
> 
> diff --git a/net/core/datagram.c b/net/core/datagram.c
> index 7f7d5da2e406..2b24d69b1e94 100644
> --- a/net/core/datagram.c
> +++ b/net/core/datagram.c
> @@ -610,7 +610,7 @@ int skb_copy_datagram_from_iter(struct sk_buff *skb, int offset,
>  }
>  EXPORT_SYMBOL(skb_copy_datagram_from_iter);
>  
> -static int zerocopy_fill_skb_from_iter(struct sock *sk, struct sk_buff *skb,
> +static int zerocopy_fill_skb_from_iter(struct sk_buff *skb,
>  					struct iov_iter *from, size_t length)
>  {
>  	int frag = skb_shinfo(skb)->nr_frags;
> @@ -621,7 +621,6 @@ static int zerocopy_fill_skb_from_iter(struct sock *sk, struct sk_buff *skb,
>  		int refs, order, n = 0;
>  		size_t start;
>  		ssize_t copied;
> -		unsigned long truesize;
>  
>  		if (frag == MAX_SKB_FRAGS)
>  			return -EMSGSIZE;

Does the existing code then incorrectly not unwind sk_wmem_queued_add
and sk_mem_charge if returning with error from the second or later
loop..

> @@ -633,17 +632,9 @@ static int zerocopy_fill_skb_from_iter(struct sock *sk, struct sk_buff *skb,
>  
>  		length -= copied;
>  
> -		truesize = PAGE_ALIGN(copied + start);
>  		skb->data_len += copied;
>  		skb->len += copied;
> -		skb->truesize += truesize;
> -		if (sk && sk->sk_type == SOCK_STREAM) {
> -			sk_wmem_queued_add(sk, truesize);
> -			if (!skb_zcopy_pure(skb))
> -				sk_mem_charge(sk, truesize);
> -		} else {
> -			refcount_add(truesize, &skb->sk->sk_wmem_alloc);
> -		}
> +		skb->truesize += PAGE_ALIGN(copied + start);
>  
>  		head = compound_head(pages[n]);
>  		order = compound_order(head);
> @@ -691,10 +682,24 @@ int __zerocopy_sg_from_iter(struct msghdr *msg, struct sock *sk,
>  			    struct sk_buff *skb, struct iov_iter *from,
>  			    size_t length)
>  {
> +	unsigned long orig_size = skb->truesize;
> +	unsigned long truesize;
> +	int ret;
> +
>  	if (msg && msg->msg_ubuf && msg->sg_from_iter)
>  		return msg->sg_from_iter(sk, skb, from, length);
> -	else
> -		return zerocopy_fill_skb_from_iter(sk, skb, from, length);
> +
> +	ret = zerocopy_fill_skb_from_iter(skb, from, length);
> +	truesize = skb->truesize - orig_size;
> +
> +	if (sk && sk->sk_type == SOCK_STREAM) {
> +		sk_wmem_queued_add(sk, truesize);
> +		if (!skb_zcopy_pure(skb))
> +			sk_mem_charge(sk, truesize);
> +	} else {
> +		refcount_add(truesize, &skb->sk->sk_wmem_alloc);
> +	}
> +	return ret;
>  }
>  EXPORT_SYMBOL(__zerocopy_sg_from_iter);
>  
> -- 
> 2.44.0
> 



  reply	other threads:[~2024-06-28 17:06 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-27 12:59 [PATCH net-next 0/5] zerocopy tx cleanups Pavel Begunkov
2024-06-27 12:59 ` [PATCH net-next 1/5] net: always try to set ubuf in skb_zerocopy_iter_stream Pavel Begunkov
2024-06-28 17:03   ` Willem de Bruijn
2024-06-27 12:59 ` [PATCH net-next 2/5] net: split __zerocopy_sg_from_iter() Pavel Begunkov
2024-06-28 17:03   ` Willem de Bruijn
2024-06-27 12:59 ` [PATCH net-next 3/5] net: batch zerocopy_fill_skb_from_iter accounting Pavel Begunkov
2024-06-28 17:06   ` Willem de Bruijn [this message]
2024-07-01 11:11     ` Pavel Begunkov
2024-07-01 18:27       ` Willem de Bruijn
2024-06-27 12:59 ` [PATCH net-next 4/5] io_uring/net: move charging socket out of zc io_uring Pavel Begunkov
2024-06-28 17:06   ` Willem de Bruijn
2024-06-27 12:59 ` [PATCH net-next 5/5] net: limit scope of a skb_zerocopy_iter_stream var Pavel Begunkov
2024-06-28 17:07   ` Willem de Bruijn
2024-06-28 17:09 ` [PATCH net-next 0/5] zerocopy tx cleanups Jens Axboe
2024-07-02 10:30 ` patchwork-bot+netdevbpf

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=667eed8350f89_2185b294e2@willemb.c.googlers.com.notmuch \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox