From: David Gibson <david@gibson.dropbear.id.au>
To: Laurent Vivier <lvivier@redhat.com>
Cc: passt-dev@passt.top
Subject: Re: [PATCH v4 02/10] tcp: extract buffer management from tcp_send_flag()
Date: Sat, 1 Jun 2024 15:43:08 +1000 [thread overview]
Message-ID: <Zlq07KjbsyFoAfJz@zatzit> (raw)
In-Reply-To: <20240531142344.1420034-3-lvivier@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 5499 bytes --]
On Fri, May 31, 2024 at 04:23:36PM +0200, Laurent Vivier wrote:
I think this needs a commit message. In particular expanding on what
you mean by "buffer management" would be helpful.
> Signed-off-by: Laurent Vivier <lvivier@redhat.com>
> ---
> tcp.c | 87 ++++++++++++++++++++++++++++++++++++-----------------------
> 1 file changed, 54 insertions(+), 33 deletions(-)
>
> diff --git a/tcp.c b/tcp.c
> index 6f221995f3bc..a6f43010f58f 100644
> --- a/tcp.c
> +++ b/tcp.c
> @@ -1518,24 +1518,25 @@ static void tcp_update_seqack_from_tap(const struct ctx *c,
> }
>
> /**
> - * tcp_send_flag() - Send segment with flags to tap (no payload)
> + * tcp_fill_flag_header() - Prepare header for flags-only segment (no payload)
> * @c: Execution context
> * @conn: Connection pointer
> * @flags: TCP flags: if not set, send segment only if ACK is due
> + * @th: TCP header to update
> + * @data: buffer to store TCP option
> + * @optlen: size of the TCP option buffer
> *
> - * Return: negative error code on connection reset, 0 otherwise
> + * Return: < 0 error code on connection reset,
> + * 0 if there is no flag to send
> + * 1 otherwise
> */
> -static int tcp_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags)
> +static int tcp_fill_flag_header(struct ctx *c, struct tcp_tap_conn *conn,
> + int flags, struct tcphdr *th, char *data,
> + size_t *optlen)
> {
> - struct tcp_flags_t *payload;
> struct tcp_info tinfo = { 0 };
> socklen_t sl = sizeof(tinfo);
> int s = conn->sock;
> - size_t optlen = 0;
> - struct tcphdr *th;
> - struct iovec *iov;
> - size_t l4len;
> - char *data;
>
> if (SEQ_GE(conn->seq_ack_to_tap, conn->seq_from_tap) &&
> !flags && conn->wnd_to_tap)
> @@ -1557,20 +1558,11 @@ static int tcp_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags)
> if (!tcp_update_seqack_wnd(c, conn, flags, &tinfo) && !flags)
> return 0;
>
> - if (CONN_V4(conn))
> - iov = tcp4_l2_flags_iov[tcp4_flags_used++];
> - else
> - iov = tcp6_l2_flags_iov[tcp6_flags_used++];
> -
> - payload = iov[TCP_IOV_PAYLOAD].iov_base;
> - th = &payload->th;
> - data = payload->opts;
> -
> if (flags & SYN) {
> int mss;
>
> /* Options: MSS, NOP and window scale (8 bytes) */
> - optlen = OPT_MSS_LEN + 1 + OPT_WS_LEN;
> + *optlen = OPT_MSS_LEN + 1 + OPT_WS_LEN;
>
> *data++ = OPT_MSS;
> *data++ = OPT_MSS_LEN;
> @@ -1604,26 +1596,13 @@ static int tcp_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags)
> flags |= ACK;
> }
>
> - th->doff = (sizeof(*th) + optlen) / 4;
> + th->doff = (sizeof(*th) + *optlen) / 4;
>
> th->ack = !!(flags & ACK);
> th->rst = !!(flags & RST);
> th->syn = !!(flags & SYN);
> th->fin = !!(flags & FIN);
>
> - if (CONN_V4(conn)) {
> - l4len = tcp_fill_headers4(c, conn, iov[TCP_IOV_TAP].iov_base,
> - iov[TCP_IOV_IP].iov_base,
> - iov[TCP_IOV_PAYLOAD].iov_base, optlen,
> - NULL, conn->seq_to_tap);
> - } else {
> - l4len = tcp_fill_headers6(c, conn, iov[TCP_IOV_TAP].iov_base,
> - iov[TCP_IOV_IP].iov_base,
> - iov[TCP_IOV_PAYLOAD].iov_base, optlen,
> - conn->seq_to_tap);
> - }
> - iov[TCP_IOV_PAYLOAD].iov_len = l4len;
> -
> if (th->ack) {
> if (SEQ_GE(conn->seq_ack_to_tap, conn->seq_from_tap))
> conn_flag(c, conn, ~ACK_TO_TAP_DUE);
> @@ -1638,6 +1617,48 @@ static int tcp_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags)
> if (th->fin || th->syn)
> conn->seq_to_tap++;
>
> + return 1;
> +}
> +
> +static int tcp_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags)
> +{
> + struct tcp_flags_t *payload;
> + size_t optlen = 0;
> + struct iovec *iov;
> + size_t l4len;
> + int ret;
> +
> + if (CONN_V4(conn)) {
> + iov = tcp4_l2_flags_iov[tcp4_flags_used++];
Hrm.. it does occur to me that if you avoided the previous patch,
keeping a single dispatcher for filling in the IP headers, then
setting the iov pointer would become the only v4/v6 dependent part of
this function, helping us down the road of unifying that more.
Obviously there would be a separate CONN_v4() check hidden inside
fill_headers(), but as I've said I'm already aiming to replace those
with more specific tests.
> + payload = iov[TCP_IOV_PAYLOAD].iov_base;
> +
> + ret = tcp_fill_flag_header(c, conn, flags, &payload->th,
> + payload->opts, &optlen);
> + if (ret <= 0)
> + return ret;
> +
> + l4len = tcp_fill_headers4(c, conn, iov[TCP_IOV_TAP].iov_base,
> + iov[TCP_IOV_IP].iov_base,
> + iov[TCP_IOV_PAYLOAD].iov_base, optlen,
> + NULL, conn->seq_to_tap);
> + } else {
> + iov = tcp6_l2_flags_iov[tcp6_flags_used++];
> +
> + payload = iov[TCP_IOV_PAYLOAD].iov_base;
> +
> + ret = tcp_fill_flag_header(c, conn, flags, &payload->th,
> + payload->opts, &optlen);
> + if (ret <= 0)
> + return ret;
> +
> + l4len = tcp_fill_headers6(c, conn, iov[TCP_IOV_TAP].iov_base,
> + iov[TCP_IOV_IP].iov_base,
> + iov[TCP_IOV_PAYLOAD].iov_base, optlen,
> + conn->seq_to_tap);
> + }
> + iov[TCP_IOV_PAYLOAD].iov_len = l4len;
> +
> if (flags & DUP_ACK) {
> struct iovec *dup_iov;
> int i;
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next prev parent reply other threads:[~2024-06-01 6:38 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-31 14:23 [PATCH v4 00/10] Add vhost-user support to passt (part 2) Laurent Vivier
2024-05-31 14:23 ` [PATCH v4 01/10] tcp: inline tcp_l2_buf_fill_headers() Laurent Vivier
2024-05-31 14:23 ` [PATCH v4 02/10] tcp: extract buffer management from tcp_send_flag() Laurent Vivier
2024-06-01 5:43 ` David Gibson [this message]
2024-05-31 14:23 ` [PATCH v4 03/10] tcp: move buffers management functions to their own file Laurent Vivier
2024-06-03 1:27 ` David Gibson
2024-05-31 14:23 ` [PATCH v4 04/10] tap: export pool_flush()/tapX_handler()/packet_add() Laurent Vivier
2024-06-03 1:32 ` David Gibson
2024-05-31 14:23 ` [PATCH v4 05/10] udp: move udpX_l2_buf_t and udpX_l2_mh_sock out of udp_update_hdrX() Laurent Vivier
2024-06-03 2:54 ` David Gibson
2024-05-31 14:23 ` [PATCH v4 06/10] udp: rename udp_sock_handler() to udp_buf_sock_handler() Laurent Vivier
2024-06-03 4:02 ` David Gibson
2024-05-31 14:23 ` [PATCH v4 07/10] vhost-user: compare mode MODE_PASTA and not MODE_PASST Laurent Vivier
2024-06-03 4:04 ` David Gibson
2024-05-31 14:23 ` [PATCH v4 08/10] iov: remove iov_copy() Laurent Vivier
2024-06-03 4:05 ` David Gibson
2024-05-31 14:23 ` [PATCH v4 09/10] tcp: remove tap_hdr parameter Laurent Vivier
2024-06-03 4:12 ` David Gibson
2024-05-31 14:23 ` [PATCH v4 10/10] tap: use in->buf_size rather than sizeof(pkt_buf) Laurent Vivier
2024-06-03 4:20 ` David Gibson
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=Zlq07KjbsyFoAfJz@zatzit \
--to=david@gibson.dropbear.id.au \
--cc=lvivier@redhat.com \
--cc=passt-dev@passt.top \
/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.
Code repositories for project(s) associated with this public inbox
https://passt.top/passt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for IMAP folder(s).