netdev - [PATCH net-next 11/11] tls: rx: jump out for cases which need to leave skb on list

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20220408183134.1054551-12-kuba@kernel.org>
Date:   Fri,  8 Apr 2022 11:31:34 -0700
From:   Jakub Kicinski <kuba@...nel.org>
To:     davem@...emloft.net, pabeni@...hat.com
Cc:     netdev@...r.kernel.org, borisp@...dia.com,
        john.fastabend@...il.com, daniel@...earbox.net,
        vfedorenko@...ek.ru, Jakub Kicinski <kuba@...nel.org>
Subject: [PATCH net-next 11/11] tls: rx: jump out for cases which need to leave skb on list

The current invese logic is harder to follow (and adds extra
tests to the fast path). We have to enumerate all cases which
need to keep the skb before consuming it. It's simpler to
jump out of the full record flow as we detect those cases.

This makes it clear that partial consumption and peek can
only reach end of the function thru the !zc case so move
the code up there.

Signed-off-by: Jakub Kicinski <kuba@...nel.org>
---
 net/tls/tls_sw.c | 43 ++++++++++++++++++++++---------------------
 1 file changed, 22 insertions(+), 21 deletions(-)

diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
index 71d8082647c8..2e8a896af81a 100644
--- a/net/tls/tls_sw.c
+++ b/net/tls/tls_sw.c
@@ -1773,7 +1773,6 @@ int tls_sw_recvmsg(struct sock *sk,
 	decrypted = 0;
 	while (len && (decrypted + copied < target || ctx->recv_pkt)) {
 		struct tls_decrypt_arg darg = {};
-		bool retain_skb = false;
 		int to_decrypt, chunk;
 
 		skb = tls_wait_data(sk, psock, flags & MSG_DONTWAIT, timeo, &err);
@@ -1833,12 +1832,17 @@ int tls_sw_recvmsg(struct sock *sk,
 		if (async) {
 			/* TLS 1.2-only, to_decrypt must be text length */
 			chunk = min_t(int, to_decrypt, len);
-			goto pick_next_record;
+leave_on_list:
+			decrypted += chunk;
+			len -= chunk;
+			continue;
 		}
 		/* TLS 1.3 may have updated the length by more than overhead */
 		chunk = rxm->full_len;
 
 		if (!darg.zc) {
+			bool partially_consumed = chunk > len;
+
 			if (bpf_strp_enabled) {
 				err = sk_psock_tls_strp_read(psock, skb);
 				if (err != __SK_PASS) {
@@ -1851,39 +1855,36 @@ int tls_sw_recvmsg(struct sock *sk,
 				}
 			}
 
-			if (chunk > len) {
-				retain_skb = true;
+			if (partially_consumed)
 				chunk = len;
-			}
 
 			err = skb_copy_datagram_msg(skb, rxm->offset,
 						    msg, chunk);
 			if (err < 0)
 				goto recv_end;
 
-			if (!is_peek) {
-				rxm->offset = rxm->offset + chunk;
-				rxm->full_len = rxm->full_len - chunk;
+			if (is_peek)
+				goto leave_on_list;
+
+			if (partially_consumed) {
+				rxm->offset += chunk;
+				rxm->full_len -= chunk;
+				goto leave_on_list;
 			}
 		}
 
-pick_next_record:
 		decrypted += chunk;
 		len -= chunk;
 
-		/* For async or peek case, queue the current skb */
-		if (!(async || is_peek || retain_skb)) {
-			skb_unlink(skb, &ctx->rx_list);
-			consume_skb(skb);
+		skb_unlink(skb, &ctx->rx_list);
+		consume_skb(skb);
 
-			/* Return full control message to
-			 * userspace before trying to parse
-			 * another message type
-			 */
-			msg->msg_flags |= MSG_EOR;
-			if (control != TLS_RECORD_TYPE_DATA)
-				goto recv_end;
-		}
+		/* Return full control message to userspace before trying
+		 * to parse another message type
+		 */
+		msg->msg_flags |= MSG_EOR;
+		if (control != TLS_RECORD_TYPE_DATA)
+			break;
 	}
 
 recv_end:
-- 
2.34.1