rxrpc: Rewrite the data and ack handling code

Rewrite the data and ack handling code such that:

 (1) Parsing of received ACK and ABORT packets and the distribution and the
     filing of DATA packets happens entirely within the data_ready context
     called from the UDP socket.  This allows us to process and discard ACK
     and ABORT packets much more quickly (they're no longer stashed on a
     queue for a background thread to process).

 (2) We avoid calling skb_clone(), pskb_pull() and pskb_trim().  We instead
     keep track of the offset and length of the content of each packet in
     the sk_buff metadata.  This means we don't do any allocation in the
     receive path.

 (3) Jumbo DATA packet parsing is now done in data_ready context.  Rather
     than cloning the packet once for each subpacket and pulling/trimming
     it, we file the packet multiple times with an annotation for each
     indicating which subpacket is there.  From that we can directly
     calculate the offset and length.

 (4) A call's receive queue can be accessed without taking locks (memory
     barriers do have to be used, though).

 (5) Incoming calls are set up from preallocated resources and immediately
     made live.  They can than have packets queued upon them and ACKs
     generated.  If insufficient resources exist, DATA packet #1 is given a
     BUSY reply and other DATA packets are discarded).

 (6) sk_buffs no longer take a ref on their parent call.

To make this work, the following changes are made:

 (1) Each call's receive buffer is now a circular buffer of sk_buff
     pointers (rxtx_buffer) rather than a number of sk_buff_heads spread
     between the call and the socket.  This permits each sk_buff to be in
     the buffer multiple times.  The receive buffer is reused for the
     transmit buffer.

 (2) A circular buffer of annotations (rxtx_annotations) is kept parallel
     to the data buffer.  Transmission phase annotations indicate whether a
     buffered packet has been ACK'd or not and whether it needs
     retransmission.

     Receive phase annotations indicate whether a slot holds a whole packet
     or a jumbo subpacket and, if the latter, which subpacket.  They also
     note whether the packet has been decrypted in place.

 (3) DATA packet window tracking is much simplified.  Each phase has just
     two numbers representing the window (rx_hard_ack/rx_top and
     tx_hard_ack/tx_top).

     The hard_ack number is the sequence number before base of the window,
     representing the last packet the other side says it has consumed.
     hard_ack starts from 0 and the first packet is sequence number 1.

     The top number is the sequence number of the highest-numbered packet
     residing in the buffer.  Packets between hard_ack+1 and top are
     soft-ACK'd to indicate they've been received, but not yet consumed.

     Four macros, before(), before_eq(), after() and after_eq() are added
     to compare sequence numbers within the window.  This allows for the
     top of the window to wrap when the hard-ack sequence number gets close
     to the limit.

     Two flags, RXRPC_CALL_RX_LAST and RXRPC_CALL_TX_LAST, are added also
     to indicate when rx_top and tx_top point at the packets with the
     LAST_PACKET bit set, indicating the end of the phase.

 (4) Calls are queued on the socket 'receive queue' rather than packets.
     This means that we don't need have to invent dummy packets to queue to
     indicate abnormal/terminal states and we don't have to keep metadata
     packets (such as ABORTs) around

 (5) The offset and length of a (sub)packet's content are now passed to
     the verify_packet security op.  This is currently expected to decrypt
     the packet in place and validate it.

     However, there's now nowhere to store the revised offset and length of
     the actual data within the decrypted blob (there may be a header and
     padding to skip) because an sk_buff may represent multiple packets, so
     a locate_data security op is added to retrieve these details from the
     sk_buff content when needed.

 (6) recvmsg() now has to handle jumbo subpackets, where each subpacket is
     individually secured and needs to be individually decrypted.  The code
     to do this is broken out into rxrpc_recvmsg_data() and shared with the
     kernel API.  It now iterates over the call's receive buffer rather
     than walking the socket receive queue.

Additional changes:

 (1) The timers are condensed to a single timer that is set for the soonest
     of three timeouts (delayed ACK generation, DATA retransmission and
     call lifespan).

 (2) Transmission of ACK and ABORT packets is effected immediately from
     process-context socket ops/kernel API calls that cause them instead of
     them being punted off to a background work item.  The data_ready
     handler still has to defer to the background, though.

 (3) A shutdown op is added to the AF_RXRPC socket so that the AFS
     filesystem can shut down the socket and flush its own work items
     before closing the socket to deal with any in-progress service calls.

Future additional changes that will need to be considered:

 (1) Make sure that a call doesn't hog the front of the queue by receiving
     data from the network as fast as userspace is consuming it to the
     exclusion of other calls.

 (2) Transmit delayed ACKs from within recvmsg() when we've consumed
     sufficiently more packets to avoid the background work item needing to
     run.

Signed-off-by: David Howells <dhowells@redhat.com>

1 files changed, 47 insertions, 79 deletions

diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c
index 9a4af992fcdf..cba236575073 100644
--- a/net/rxrpc/sendmsg.c
+++ b/net/rxrpc/sendmsg.c
@@ -15,7 +15,6 @@
 #include <linux/gfp.h>
 #include <linux/skbuff.h>
 #include <linux/export.h>
-#include <linux/circ_buf.h>
 #include <net/sock.h>
 #include <net/af_rxrpc.h>
 #include "ar-internal.h"
@@ -38,19 +37,20 @@ static int rxrpc_wait_for_tx_window(struct rxrpc_sock *rx,
 	DECLARE_WAITQUEUE(myself, current);
 	int ret;
 
-	_enter(",{%d},%ld",
-	       CIRC_SPACE(call->acks_head, ACCESS_ONCE(call->acks_tail),
-			  call->acks_winsz),
-	       *timeo);
+	_enter(",{%u,%u,%u}",
+	       call->tx_hard_ack, call->tx_top, call->tx_winsize);
 
 	add_wait_queue(&call->waitq, &myself);
 
 	for (;;) {
 		set_current_state(TASK_INTERRUPTIBLE);
 		ret = 0;
-		if (CIRC_SPACE(call->acks_head, ACCESS_ONCE(call->acks_tail),
-			       call->acks_winsz) > 0)
+		if (call->tx_top - call->tx_hard_ack < call->tx_winsize)
 			break;
+		if (call->state >= RXRPC_CALL_COMPLETE) {
+			ret = -call->error;
+			break;
+		}
 		if (signal_pending(current)) {
 			ret = sock_intr_errno(*timeo);
 			break;
@@ -68,36 +68,44 @@ static int rxrpc_wait_for_tx_window(struct rxrpc_sock *rx,
 }
 
 /*
- * attempt to schedule an instant Tx resend
+ * Schedule an instant Tx resend.
  */
-static inline void rxrpc_instant_resend(struct rxrpc_call *call)
+static inline void rxrpc_instant_resend(struct rxrpc_call *call, int ix)
 {
-	read_lock_bh(&call->state_lock);
-	if (try_to_del_timer_sync(&call->resend_timer) >= 0) {
-		clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags);
-		if (call->state < RXRPC_CALL_COMPLETE &&
-		    !test_and_set_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events))
+	spin_lock_bh(&call->lock);
+
+	if (call->state < RXRPC_CALL_COMPLETE) {
+		call->rxtx_annotations[ix] = RXRPC_TX_ANNO_RETRANS;
+		if (!test_and_set_bit(RXRPC_CALL_EV_RESEND, &call->events))
 			rxrpc_queue_call(call);
 	}
-	read_unlock_bh(&call->state_lock);
+
+	spin_unlock_bh(&call->lock);
 }
 
 /*
- * queue a packet for transmission, set the resend timer and attempt
- * to send the packet immediately
+ * Queue a DATA packet for transmission, set the resend timeout and send the
+ * packet immediately
  */
 static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb,
 			       bool last)
 {
 	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
-	int ret;
+	rxrpc_seq_t seq = sp->hdr.seq;
+	int ret, ix;
+
+	_net("queue skb %p [%d]", skb, seq);
 
-	_net("queue skb %p [%d]", skb, call->acks_head);
+	ASSERTCMP(seq, ==, call->tx_top + 1);
 
-	ASSERT(call->acks_window != NULL);
-	call->acks_window[call->acks_head] = (unsigned long) skb;
+	ix = seq & RXRPC_RXTX_BUFF_MASK;
+	rxrpc_get_skb(skb);
+	call->rxtx_annotations[ix] = RXRPC_TX_ANNO_UNACK;
 	smp_wmb();
-	call->acks_head = (call->acks_head + 1) & (call->acks_winsz - 1);
+	call->rxtx_buffer[ix] = skb;
+	call->tx_top = seq;
+	if (last)
+		set_bit(RXRPC_CALL_TX_LAST, &call->flags);
 
 	if (last || call->state == RXRPC_CALL_SERVER_ACK_REQUEST) {
 		_debug("________awaiting reply/ACK__________");
@@ -121,34 +129,17 @@ static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb,
 
 	_proto("Tx DATA %%%u { #%u }", sp->hdr.serial, sp->hdr.seq);
 
-	sp->need_resend = false;
-	sp->resend_at = jiffies + rxrpc_resend_timeout;
-	if (!test_and_set_bit(RXRPC_CALL_RUN_RTIMER, &call->flags)) {
-		_debug("run timer");
-		call->resend_timer.expires = sp->resend_at;
-		add_timer(&call->resend_timer);
-	}
-
-	/* attempt to cancel the rx-ACK timer, deferring reply transmission if
-	 * we're ACK'ing the request phase of an incoming call */
-	ret = -EAGAIN;
-	if (try_to_del_timer_sync(&call->ack_timer) >= 0) {
-		/* the packet may be freed by rxrpc_process_call() before this
-		 * returns */
-		if (rxrpc_is_client_call(call))
-			rxrpc_expose_client_call(call);
-		ret = rxrpc_send_data_packet(call->conn, skb);
-		_net("sent skb %p", skb);
-	} else {
-		_debug("failed to delete ACK timer");
-	}
+	if (seq == 1 && rxrpc_is_client_call(call))
+		rxrpc_expose_client_call(call);
 
+	sp->resend_at = jiffies + rxrpc_resend_timeout;
+	ret = rxrpc_send_data_packet(call->conn, skb);
 	if (ret < 0) {
 		_debug("need instant resend %d", ret);
-		sp->need_resend = true;
-		rxrpc_instant_resend(call);
+		rxrpc_instant_resend(call, ix);
 	}
 
+	rxrpc_free_skb(skb);
 	_leave("");
 }
 
@@ -212,9 +203,8 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
 
 			_debug("alloc");
 
-			if (CIRC_SPACE(call->acks_head,
-				       ACCESS_ONCE(call->acks_tail),
-				       call->acks_winsz) <= 0) {
+			if (call->tx_top - call->tx_hard_ack >=
+			    call->tx_winsize) {
 				ret = -EAGAIN;
 				if (msg->msg_flags & MSG_DONTWAIT)
 					goto maybe_error;
@@ -313,7 +303,7 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
 					memset(skb_put(skb, pad), 0, pad);
 			}
 
-			seq = atomic_inc_return(&call->sequence);
+			seq = call->tx_top + 1;
 
 			sp->hdr.epoch	= conn->proto.epoch;
 			sp->hdr.cid	= call->cid;
@@ -329,9 +319,8 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
 			sp->hdr.flags = conn->out_clientflag;
 			if (msg_data_left(msg) == 0 && !more)
 				sp->hdr.flags |= RXRPC_LAST_PACKET;
-			else if (CIRC_SPACE(call->acks_head,
-					    ACCESS_ONCE(call->acks_tail),
-					    call->acks_winsz) > 1)
+			else if (call->tx_top - call->tx_hard_ack <
+				 call->tx_winsize)
 				sp->hdr.flags |= RXRPC_MORE_PACKETS;
 			if (more && seq & 1)
 				sp->hdr.flags |= RXRPC_REQUEST_ACK;
@@ -358,7 +347,7 @@ out:
 call_terminated:
 	rxrpc_free_skb(skb);
 	_leave(" = %d", -call->error);
-	return ret;
+	return -call->error;
 
 maybe_error:
 	if (copied)
@@ -452,29 +441,6 @@ static int rxrpc_sendmsg_cmsg(struct msghdr *msg,
 }
 
 /*
- * abort a call, sending an ABORT packet to the peer
- */
-static void rxrpc_send_abort(struct rxrpc_call *call, const char *why,
-			     u32 abort_code, int error)
-{
-	if (call->state >= RXRPC_CALL_COMPLETE)
-		return;
-
-	write_lock_bh(&call->state_lock);
-
-	if (__rxrpc_abort_call(why, call, 0, abort_code, error)) {
-		del_timer_sync(&call->resend_timer);
-		del_timer_sync(&call->ack_timer);
-		clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events);
-		clear_bit(RXRPC_CALL_EV_ACK, &call->events);
-		clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags);
-		rxrpc_queue_call(call);
-	}
-
-	write_unlock_bh(&call->state_lock);
-}
-
-/*
  * Create a new client call for sendmsg().
  */
 static struct rxrpc_call *
@@ -549,7 +515,6 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len)
 			return PTR_ERR(call);
 	}
 
-	rxrpc_see_call(call);
 	_debug("CALL %d USR %lx ST %d on CONN %p",
 	       call->debug_id, call->user_call_ID, call->state, call->conn);
 
@@ -557,8 +522,10 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len)
 		/* it's too late for this call */
 		ret = -ESHUTDOWN;
 	} else if (cmd == RXRPC_CMD_SEND_ABORT) {
-		rxrpc_send_abort(call, "CMD", abort_code, ECONNABORTED);
 		ret = 0;
+		if (rxrpc_abort_call("CMD", call, 0, abort_code, ECONNABORTED))
+			ret = rxrpc_send_call_packet(call,
+						     RXRPC_PACKET_TYPE_ABORT);
 	} else if (cmd != RXRPC_CMD_SEND_DATA) {
 		ret = -EINVAL;
 	} else if (rxrpc_is_client_call(call) &&
@@ -639,7 +606,8 @@ void rxrpc_kernel_abort_call(struct socket *sock, struct rxrpc_call *call,
 
 	lock_sock(sock->sk);
 
-	rxrpc_send_abort(call, why, abort_code, error);
+	if (rxrpc_abort_call(why, call, 0, abort_code, error))
+		rxrpc_send_call_packet(call, RXRPC_PACKET_TYPE_ABORT);
 
 	release_sock(sock->sk);
 	_leave("");