All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jakub Kicinski <kuba@kernel.org>
To: davem@davemloft.net
Cc: netdev@vger.kernel.org, edumazet@google.com, pabeni@redhat.com,
	borisp@nvidia.com, john.fastabend@gmail.com, maximmi@nvidia.com,
	tariqt@nvidia.com, vfedorenko@novek.ru,
	Jakub Kicinski <kuba@kernel.org>
Subject: [PATCH net-next v2 02/11] tls: rx: don't try to keep the skbs always on the list
Date: Thu, 14 Jul 2022 22:22:26 -0700	[thread overview]
Message-ID: <20220715052235.1452170-3-kuba@kernel.org> (raw)
In-Reply-To: <20220715052235.1452170-1-kuba@kernel.org>

I thought that having the skb either always on the ctx->rx_list
or ctx->recv_pkt will simplify the handling, as we would not
have to remember to flip it from one to the other on exit paths.

This became a little harder to justify with the fix for BPF
sockmaps. Subsequent changes will make the situation even worse.
Queue the skbs only when really needed.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/tls/tls_sw.c | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
index 761a63751616..acf65992aaca 100644
--- a/net/tls/tls_sw.c
+++ b/net/tls/tls_sw.c
@@ -1861,8 +1861,11 @@ int tls_sw_recvmsg(struct sock *sk,
 			if (psock) {
 				chunk = sk_msg_recvmsg(sk, psock, msg, len,
 						       flags);
-				if (chunk > 0)
-					goto leave_on_list;
+				if (chunk > 0) {
+					decrypted += chunk;
+					len -= chunk;
+					continue;
+				}
 			}
 			goto recv_end;
 		}
@@ -1908,14 +1911,14 @@ int tls_sw_recvmsg(struct sock *sk,
 
 		ctx->recv_pkt = NULL;
 		__strp_unpause(&ctx->strp);
-		__skb_queue_tail(&ctx->rx_list, skb);
 
 		if (async) {
 			/* TLS 1.2-only, to_decrypt must be text length */
 			chunk = min_t(int, to_decrypt, len);
-leave_on_list:
+put_on_rx_list:
 			decrypted += chunk;
 			len -= chunk;
+			__skb_queue_tail(&ctx->rx_list, skb);
 			continue;
 		}
 		/* TLS 1.3 may have updated the length by more than overhead */
@@ -1925,8 +1928,6 @@ int tls_sw_recvmsg(struct sock *sk,
 			bool partially_consumed = chunk > len;
 
 			if (bpf_strp_enabled) {
-				/* BPF may try to queue the skb */
-				__skb_unlink(skb, &ctx->rx_list);
 				err = sk_psock_tls_strp_read(psock, skb);
 				if (err != __SK_PASS) {
 					rxm->offset = rxm->offset + rxm->full_len;
@@ -1935,7 +1936,6 @@ int tls_sw_recvmsg(struct sock *sk,
 						consume_skb(skb);
 					continue;
 				}
-				__skb_queue_tail(&ctx->rx_list, skb);
 			}
 
 			if (partially_consumed)
@@ -1943,23 +1943,24 @@ int tls_sw_recvmsg(struct sock *sk,
 
 			err = skb_copy_datagram_msg(skb, rxm->offset,
 						    msg, chunk);
-			if (err < 0)
+			if (err < 0) {
+				__skb_queue_tail(&ctx->rx_list, skb);
 				goto recv_end;
+			}
 
 			if (is_peek)
-				goto leave_on_list;
+				goto put_on_rx_list;
 
 			if (partially_consumed) {
 				rxm->offset += chunk;
 				rxm->full_len -= chunk;
-				goto leave_on_list;
+				goto put_on_rx_list;
 			}
 		}
 
 		decrypted += chunk;
 		len -= chunk;
 
-		__skb_unlink(skb, &ctx->rx_list);
 		consume_skb(skb);
 
 		/* Return full control message to userspace before trying
-- 
2.36.1


  parent reply	other threads:[~2022-07-15  5:22 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-07-15  5:22 [PATCH net-next v2 00/11] tls: rx: avoid skb_cow_data() Jakub Kicinski
2022-07-15  5:22 ` [PATCH net-next v2 01/11] tls: rx: allow only one reader at a time Jakub Kicinski
2022-07-20  8:37   ` Eric Dumazet
2022-07-20 16:59     ` Jakub Kicinski
2022-07-20 17:09       ` Eric Dumazet
2022-07-20 17:19         ` Eric Dumazet
2022-07-15  5:22 ` Jakub Kicinski [this message]
2022-07-15  5:22 ` [PATCH net-next v2 03/11] tls: rx: don't keep decrypted skbs on ctx->recv_pkt Jakub Kicinski
2022-07-15  5:22 ` [PATCH net-next v2 04/11] tls: rx: remove the message decrypted tracking Jakub Kicinski
2022-07-15  5:22 ` [PATCH net-next v2 05/11] tls: rx: factor out device darg update Jakub Kicinski
2022-07-15  5:22 ` [PATCH net-next v2 06/11] tls: rx: read the input skb from ctx->recv_pkt Jakub Kicinski
2022-07-15  5:22 ` [PATCH net-next v2 07/11] tls: rx: return the decrypted skb via darg Jakub Kicinski
2022-07-15  5:22 ` [PATCH net-next v2 08/11] tls: rx: async: adjust record geometry immediately Jakub Kicinski
2022-07-15  5:22 ` [PATCH net-next v2 09/11] tls: rx: async: hold onto the input skb Jakub Kicinski
2022-07-15  5:22 ` [PATCH net-next v2 10/11] tls: rx: async: don't put async zc on the list Jakub Kicinski
2022-07-15  5:22 ` [PATCH net-next v2 11/11] tls: rx: decrypt into a fresh skb Jakub Kicinski
2022-07-18 10:40 ` [PATCH net-next v2 00/11] tls: rx: avoid skb_cow_data() 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=20220715052235.1452170-3-kuba@kernel.org \
    --to=kuba@kernel.org \
    --cc=borisp@nvidia.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=john.fastabend@gmail.com \
    --cc=maximmi@nvidia.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=tariqt@nvidia.com \
    --cc=vfedorenko@novek.ru \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.