linux-stable/net/rxrpc/proc.c

576 lines
16 KiB
C
Raw Permalink Normal View History

// SPDX-License-Identifier: GPL-2.0-or-later
/* /proc/net/ support for AF_RXRPC
*
* Copyright (C) 2007 Red Hat, Inc. All Rights Reserved.
* Written by David Howells (dhowells@redhat.com)
*/
#include <linux/module.h>
#include <net/sock.h>
#include <net/af_rxrpc.h>
#include "ar-internal.h"
static const char *const rxrpc_conn_states[RXRPC_CONN__NR_STATES] = {
[RXRPC_CONN_UNUSED] = "Unused ",
[RXRPC_CONN_CLIENT_UNSECURED] = "ClUnsec ",
[RXRPC_CONN_CLIENT] = "Client ",
rxrpc: Preallocate peers, conns and calls for incoming service requests Make it possible for the data_ready handler called from the UDP transport socket to completely instantiate an rxrpc_call structure and make it immediately live by preallocating all the memory it might need. The idea is to cut out the background thread usage as much as possible. [Note that the preallocated structs are not actually used in this patch - that will be done in a future patch.] If insufficient resources are available in the preallocation buffers, it will be possible to discard the DATA packet in the data_ready handler or schedule a BUSY packet without the need to schedule an attempt at allocation in a background thread. To this end: (1) Preallocate rxrpc_peer, rxrpc_connection and rxrpc_call structs to a maximum number each of the listen backlog size. The backlog size is limited to a maxmimum of 32. Only this many of each can be in the preallocation buffer. (2) For userspace sockets, the preallocation is charged initially by listen() and will be recharged by accepting or rejecting pending new incoming calls. (3) For kernel services {,re,dis}charging of the preallocation buffers is handled manually. Two notifier callbacks have to be provided before kernel_listen() is invoked: (a) An indication that a new call has been instantiated. This can be used to trigger background recharging. (b) An indication that a call is being discarded. This is used when the socket is being released. A function, rxrpc_kernel_charge_accept() is called by the kernel service to preallocate a single call. It should be passed the user ID to be used for that call and a callback to associate the rxrpc call with the kernel service's side of the ID. (4) Discard the preallocation when the socket is closed. (5) Temporarily bump the refcount on the call allocated in rxrpc_incoming_call() so that rxrpc_release_call() can ditch the preallocation ref on service calls unconditionally. This will no longer be necessary once the preallocation is used. Note that this does not yet control the number of active service calls on a client - that will come in a later patch. A future development would be to provide a setsockopt() call that allows a userspace server to manually charge the preallocation buffer. This would allow user call IDs to be provided in advance and the awkward manual accept stage to be bypassed. Signed-off-by: David Howells <dhowells@redhat.com>
2016-09-08 10:10:12 +00:00
[RXRPC_CONN_SERVICE_PREALLOC] = "SvPrealc",
[RXRPC_CONN_SERVICE_UNSECURED] = "SvUnsec ",
[RXRPC_CONN_SERVICE_CHALLENGING] = "SvChall ",
[RXRPC_CONN_SERVICE] = "SvSecure",
[RXRPC_CONN_ABORTED] = "Aborted ",
};
/*
* generate a list of extant and dead calls in /proc/net/rxrpc_calls
*/
static void *rxrpc_call_seq_start(struct seq_file *seq, loff_t *_pos)
__acquires(rcu)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
rxrpc: Calls shouldn't hold socket refs rxrpc calls shouldn't hold refs on the sock struct. This was done so that the socket wouldn't go away whilst the call was in progress, such that the call could reach the socket's queues. However, we can mark the socket as requiring an RCU release and rely on the RCU read lock. To make this work, we do: (1) rxrpc_release_call() removes the call's call user ID. This is now only called from socket operations and not from the call processor: rxrpc_accept_call() / rxrpc_kernel_accept_call() rxrpc_reject_call() / rxrpc_kernel_reject_call() rxrpc_kernel_end_call() rxrpc_release_calls_on_socket() rxrpc_recvmsg() Though it is also called in the cleanup path of rxrpc_accept_incoming_call() before we assign a user ID. (2) Pass the socket pointer into rxrpc_release_call() rather than getting it from the call so that we can get rid of uninitialised calls. (3) Fix call processor queueing to pass a ref to the work queue and to release that ref at the end of the processor function (or to pass it back to the work queue if we have to requeue). (4) Skip out of the call processor function asap if the call is complete and don't requeue it if the call is complete. (5) Clean up the call immediately that the refcount reaches 0 rather than trying to defer it. Actual deallocation is deferred to RCU, however. (6) Don't hold socket refs for allocated calls. (7) Use the RCU read lock when queueing a message on a socket and treat the call's socket pointer according to RCU rules and check it for NULL. We also need to use the RCU read lock when viewing a call through procfs. (8) Transmit the final ACK/ABORT to a client call in rxrpc_release_call() if this hasn't been done yet so that we can then disconnect the call. Once the call is disconnected, it won't have any access to the connection struct and the UDP socket for the call work processor to be able to send the ACK. Terminal retransmission will be handled by the connection processor. (9) Release all calls immediately on the closing of a socket rather than trying to defer this. Incomplete calls will be aborted. The call refcount model is much simplified. Refs are held on the call by: (1) A socket's user ID tree. (2) A socket's incoming call secureq and acceptq. (3) A kernel service that has a call in progress. (4) A queued call work processor. We have to take care to put any call that we failed to queue. (5) sk_buffs on a socket's receive queue. A future patch will get rid of this. Whilst we're at it, we can do: (1) Get rid of the RXRPC_CALL_EV_RELEASE event. Release is now done entirely from the socket routines and never from the call's processor. (2) Get rid of the RXRPC_CALL_DEAD state. Calls now end in the RXRPC_CALL_COMPLETE state. (3) Get rid of the rxrpc_call::destroyer work item. Calls are now torn down when their refcount reaches 0 and then handed over to RCU for final cleanup. (4) Get rid of the rxrpc_call::deadspan timer. Calls are cleaned up immediately they're finished with and don't hang around. Post-completion retransmission is handled by the connection processor once the call is disconnected. (5) Get rid of the dead call expiry setting as there's no longer a timer to set. (6) rxrpc_destroy_all_calls() can just check that the call list is empty. Signed-off-by: David Howells <dhowells@redhat.com>
2016-09-07 08:19:31 +00:00
rcu_read_lock();
rxrpc: Fix locking issue There's a locking issue with the per-netns list of calls in rxrpc. The pieces of code that add and remove a call from the list use write_lock() and the calls procfile uses read_lock() to access it. However, the timer callback function may trigger a removal by trying to queue a call for processing and finding that it's already queued - at which point it has a spare refcount that it has to do something with. Unfortunately, if it puts the call and this reduces the refcount to 0, the call will be removed from the list. Unfortunately, since the _bh variants of the locking functions aren't used, this can deadlock. ================================ WARNING: inconsistent lock state 5.18.0-rc3-build4+ #10 Not tainted -------------------------------- inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage. ksoftirqd/2/25 [HC0[0]:SC1[1]:HE1:SE0] takes: ffff888107ac4038 (&rxnet->call_lock){+.?.}-{2:2}, at: rxrpc_put_call+0x103/0x14b {SOFTIRQ-ON-W} state was registered at: ... Possible unsafe locking scenario: CPU0 ---- lock(&rxnet->call_lock); <Interrupt> lock(&rxnet->call_lock); *** DEADLOCK *** 1 lock held by ksoftirqd/2/25: #0: ffff8881008ffdb0 ((&call->timer)){+.-.}-{0:0}, at: call_timer_fn+0x5/0x23d Changes ======= ver #2) - Changed to using list_next_rcu() rather than rcu_dereference() directly. Fixes: 17926a79320a ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both") Signed-off-by: David Howells <dhowells@redhat.com> cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-21 07:45:28 +00:00
return seq_list_start_head_rcu(&rxnet->calls, *_pos);
}
static void *rxrpc_call_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
rxrpc: Fix locking issue There's a locking issue with the per-netns list of calls in rxrpc. The pieces of code that add and remove a call from the list use write_lock() and the calls procfile uses read_lock() to access it. However, the timer callback function may trigger a removal by trying to queue a call for processing and finding that it's already queued - at which point it has a spare refcount that it has to do something with. Unfortunately, if it puts the call and this reduces the refcount to 0, the call will be removed from the list. Unfortunately, since the _bh variants of the locking functions aren't used, this can deadlock. ================================ WARNING: inconsistent lock state 5.18.0-rc3-build4+ #10 Not tainted -------------------------------- inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage. ksoftirqd/2/25 [HC0[0]:SC1[1]:HE1:SE0] takes: ffff888107ac4038 (&rxnet->call_lock){+.?.}-{2:2}, at: rxrpc_put_call+0x103/0x14b {SOFTIRQ-ON-W} state was registered at: ... Possible unsafe locking scenario: CPU0 ---- lock(&rxnet->call_lock); <Interrupt> lock(&rxnet->call_lock); *** DEADLOCK *** 1 lock held by ksoftirqd/2/25: #0: ffff8881008ffdb0 ((&call->timer)){+.-.}-{0:0}, at: call_timer_fn+0x5/0x23d Changes ======= ver #2) - Changed to using list_next_rcu() rather than rcu_dereference() directly. Fixes: 17926a79320a ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both") Signed-off-by: David Howells <dhowells@redhat.com> cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-21 07:45:28 +00:00
return seq_list_next_rcu(v, &rxnet->calls, pos);
}
static void rxrpc_call_seq_stop(struct seq_file *seq, void *v)
__releases(rcu)
{
rxrpc: Calls shouldn't hold socket refs rxrpc calls shouldn't hold refs on the sock struct. This was done so that the socket wouldn't go away whilst the call was in progress, such that the call could reach the socket's queues. However, we can mark the socket as requiring an RCU release and rely on the RCU read lock. To make this work, we do: (1) rxrpc_release_call() removes the call's call user ID. This is now only called from socket operations and not from the call processor: rxrpc_accept_call() / rxrpc_kernel_accept_call() rxrpc_reject_call() / rxrpc_kernel_reject_call() rxrpc_kernel_end_call() rxrpc_release_calls_on_socket() rxrpc_recvmsg() Though it is also called in the cleanup path of rxrpc_accept_incoming_call() before we assign a user ID. (2) Pass the socket pointer into rxrpc_release_call() rather than getting it from the call so that we can get rid of uninitialised calls. (3) Fix call processor queueing to pass a ref to the work queue and to release that ref at the end of the processor function (or to pass it back to the work queue if we have to requeue). (4) Skip out of the call processor function asap if the call is complete and don't requeue it if the call is complete. (5) Clean up the call immediately that the refcount reaches 0 rather than trying to defer it. Actual deallocation is deferred to RCU, however. (6) Don't hold socket refs for allocated calls. (7) Use the RCU read lock when queueing a message on a socket and treat the call's socket pointer according to RCU rules and check it for NULL. We also need to use the RCU read lock when viewing a call through procfs. (8) Transmit the final ACK/ABORT to a client call in rxrpc_release_call() if this hasn't been done yet so that we can then disconnect the call. Once the call is disconnected, it won't have any access to the connection struct and the UDP socket for the call work processor to be able to send the ACK. Terminal retransmission will be handled by the connection processor. (9) Release all calls immediately on the closing of a socket rather than trying to defer this. Incomplete calls will be aborted. The call refcount model is much simplified. Refs are held on the call by: (1) A socket's user ID tree. (2) A socket's incoming call secureq and acceptq. (3) A kernel service that has a call in progress. (4) A queued call work processor. We have to take care to put any call that we failed to queue. (5) sk_buffs on a socket's receive queue. A future patch will get rid of this. Whilst we're at it, we can do: (1) Get rid of the RXRPC_CALL_EV_RELEASE event. Release is now done entirely from the socket routines and never from the call's processor. (2) Get rid of the RXRPC_CALL_DEAD state. Calls now end in the RXRPC_CALL_COMPLETE state. (3) Get rid of the rxrpc_call::destroyer work item. Calls are now torn down when their refcount reaches 0 and then handed over to RCU for final cleanup. (4) Get rid of the rxrpc_call::deadspan timer. Calls are cleaned up immediately they're finished with and don't hang around. Post-completion retransmission is handled by the connection processor once the call is disconnected. (5) Get rid of the dead call expiry setting as there's no longer a timer to set. (6) rxrpc_destroy_all_calls() can just check that the call list is empty. Signed-off-by: David Howells <dhowells@redhat.com>
2016-09-07 08:19:31 +00:00
rcu_read_unlock();
}
static int rxrpc_call_seq_show(struct seq_file *seq, void *v)
{
struct rxrpc_local *local;
struct rxrpc_call *call;
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
enum rxrpc_call_state state;
rxrpc: Don't use a ring buffer for call Tx queue Change the way the Tx queueing works to make the following ends easier to achieve: (1) The filling of packets, the encryption of packets and the transmission of packets can be handled in parallel by separate threads, rather than rxrpc_sendmsg() allocating, filling, encrypting and transmitting each packet before moving onto the next one. (2) Get rid of the fixed-size ring which sets a hard limit on the number of packets that can be retained in the ring. This allows the number of packets to increase without having to allocate a very large ring or having variable-sized rings. [Note: the downside of this is that it's then less efficient to locate a packet for retransmission as we then have to step through a list and examine each buffer in the list.] (3) Allow the filler/encrypter to run ahead of the transmission window. (4) Make it easier to do zero copy UDP from the packet buffers. (5) Make it easier to do zero copy from userspace to the packet buffers - and thence to UDP (only if for unauthenticated connections). To that end, the following changes are made: (1) Use the new rxrpc_txbuf struct instead of sk_buff for keeping packets to be transmitted in. This allows them to be placed on multiple queues simultaneously. An sk_buff isn't really necessary as it's never passed on to lower-level networking code. (2) Keep the transmissable packets in a linked list on the call struct rather than in a ring. As a consequence, the annotation buffer isn't used either; rather a flag is set on the packet to indicate ackedness. (3) Use the RXRPC_CALL_TX_LAST flag to indicate that the last packet to be transmitted has been queued. Add RXRPC_CALL_TX_ALL_ACKED to indicate that all packets up to and including the last got hard acked. (4) Wire headers are now stored in the txbuf rather than being concocted on the stack and they're stored immediately before the data, thereby allowing zerocopy of a single span. (5) Don't bother with instant-resend on transmission failure; rather, leave it for a timer or an ACK packet to trigger. Signed-off-by: David Howells <dhowells@redhat.com> cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org
2022-03-31 22:55:08 +00:00
rxrpc_seq_t acks_hard_ack;
char lbuff[50], rbuff[50];
long timeout = 0;
if (v == &rxnet->calls) {
seq_puts(seq,
"Proto Local "
" Remote "
" SvID ConnID CallID End Use State Abort "
" DebugId TxSeq TW RxSeq RW RxSerial CW RxTimo\n");
return 0;
}
call = list_entry(v, struct rxrpc_call, link);
local = call->local;
if (local)
sprintf(lbuff, "%pISpc", &local->srx.transport);
else
strcpy(lbuff, "no_local");
sprintf(rbuff, "%pISpc", &call->dest_srx.transport);
state = rxrpc_call_state(call);
if (state != RXRPC_CALL_SERVER_PREALLOC)
timeout = ktime_ms_delta(READ_ONCE(call->expect_rx_by), ktime_get_real());
rxrpc: Don't use a ring buffer for call Tx queue Change the way the Tx queueing works to make the following ends easier to achieve: (1) The filling of packets, the encryption of packets and the transmission of packets can be handled in parallel by separate threads, rather than rxrpc_sendmsg() allocating, filling, encrypting and transmitting each packet before moving onto the next one. (2) Get rid of the fixed-size ring which sets a hard limit on the number of packets that can be retained in the ring. This allows the number of packets to increase without having to allocate a very large ring or having variable-sized rings. [Note: the downside of this is that it's then less efficient to locate a packet for retransmission as we then have to step through a list and examine each buffer in the list.] (3) Allow the filler/encrypter to run ahead of the transmission window. (4) Make it easier to do zero copy UDP from the packet buffers. (5) Make it easier to do zero copy from userspace to the packet buffers - and thence to UDP (only if for unauthenticated connections). To that end, the following changes are made: (1) Use the new rxrpc_txbuf struct instead of sk_buff for keeping packets to be transmitted in. This allows them to be placed on multiple queues simultaneously. An sk_buff isn't really necessary as it's never passed on to lower-level networking code. (2) Keep the transmissable packets in a linked list on the call struct rather than in a ring. As a consequence, the annotation buffer isn't used either; rather a flag is set on the packet to indicate ackedness. (3) Use the RXRPC_CALL_TX_LAST flag to indicate that the last packet to be transmitted has been queued. Add RXRPC_CALL_TX_ALL_ACKED to indicate that all packets up to and including the last got hard acked. (4) Wire headers are now stored in the txbuf rather than being concocted on the stack and they're stored immediately before the data, thereby allowing zerocopy of a single span. (5) Don't bother with instant-resend on transmission failure; rather, leave it for a timer or an ACK packet to trigger. Signed-off-by: David Howells <dhowells@redhat.com> cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org
2022-03-31 22:55:08 +00:00
acks_hard_ack = READ_ONCE(call->acks_hard_ack);
seq_printf(seq,
"UDP %-47.47s %-47.47s %4x %08x %08x %s %3u"
" %-8.8s %08x %08x %08x %02x %08x %02x %08x %02x %06lx\n",
lbuff,
rbuff,
call->dest_srx.srx_service,
call->cid,
call->call_id,
rxrpc_is_service_call(call) ? "Svc" : "Clt",
refcount_read(&call->ref),
rxrpc_call_states[state],
call->abort_code,
call->debug_id,
rxrpc: Don't use a ring buffer for call Tx queue Change the way the Tx queueing works to make the following ends easier to achieve: (1) The filling of packets, the encryption of packets and the transmission of packets can be handled in parallel by separate threads, rather than rxrpc_sendmsg() allocating, filling, encrypting and transmitting each packet before moving onto the next one. (2) Get rid of the fixed-size ring which sets a hard limit on the number of packets that can be retained in the ring. This allows the number of packets to increase without having to allocate a very large ring or having variable-sized rings. [Note: the downside of this is that it's then less efficient to locate a packet for retransmission as we then have to step through a list and examine each buffer in the list.] (3) Allow the filler/encrypter to run ahead of the transmission window. (4) Make it easier to do zero copy UDP from the packet buffers. (5) Make it easier to do zero copy from userspace to the packet buffers - and thence to UDP (only if for unauthenticated connections). To that end, the following changes are made: (1) Use the new rxrpc_txbuf struct instead of sk_buff for keeping packets to be transmitted in. This allows them to be placed on multiple queues simultaneously. An sk_buff isn't really necessary as it's never passed on to lower-level networking code. (2) Keep the transmissable packets in a linked list on the call struct rather than in a ring. As a consequence, the annotation buffer isn't used either; rather a flag is set on the packet to indicate ackedness. (3) Use the RXRPC_CALL_TX_LAST flag to indicate that the last packet to be transmitted has been queued. Add RXRPC_CALL_TX_ALL_ACKED to indicate that all packets up to and including the last got hard acked. (4) Wire headers are now stored in the txbuf rather than being concocted on the stack and they're stored immediately before the data, thereby allowing zerocopy of a single span. (5) Don't bother with instant-resend on transmission failure; rather, leave it for a timer or an ACK packet to trigger. Signed-off-by: David Howells <dhowells@redhat.com> cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org
2022-03-31 22:55:08 +00:00
acks_hard_ack, READ_ONCE(call->tx_top) - acks_hard_ack,
call->ackr_window, call->ackr_wtop - call->ackr_window,
call->rx_serial,
call->cong_cwnd,
timeout);
return 0;
}
const struct seq_operations rxrpc_call_seq_ops = {
.start = rxrpc_call_seq_start,
.next = rxrpc_call_seq_next,
.stop = rxrpc_call_seq_stop,
.show = rxrpc_call_seq_show,
};
/*
* generate a list of extant virtual connections in /proc/net/rxrpc_conns
*/
static void *rxrpc_connection_seq_start(struct seq_file *seq, loff_t *_pos)
__acquires(rxnet->conn_lock)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
read_lock(&rxnet->conn_lock);
return seq_list_start_head(&rxnet->conn_proc_list, *_pos);
}
static void *rxrpc_connection_seq_next(struct seq_file *seq, void *v,
loff_t *pos)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
return seq_list_next(v, &rxnet->conn_proc_list, pos);
}
static void rxrpc_connection_seq_stop(struct seq_file *seq, void *v)
__releases(rxnet->conn_lock)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
read_unlock(&rxnet->conn_lock);
}
static int rxrpc_connection_seq_show(struct seq_file *seq, void *v)
{
struct rxrpc_connection *conn;
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
const char *state;
char lbuff[50], rbuff[50];
if (v == &rxnet->conn_proc_list) {
seq_puts(seq,
"Proto Local "
" Remote "
rxrpc: Don't hold a ref for connection workqueue Currently, rxrpc gives the connection's work item a ref on the connection when it queues it - and this is called from the timer expiration function. The problem comes when queue_work() fails (ie. the work item is already queued): the timer routine must put the ref - but this may cause the cleanup code to run. This has the unfortunate effect that the cleanup code may then be run in softirq context - which means that any spinlocks it might need to touch have to be guarded to disable softirqs (ie. they need a "_bh" suffix). (1) Don't give a ref to the work item. (2) Simplify handling of service connections by adding a separate active count so that the refcount isn't also used for this. (3) Connection destruction for both client and service connections can then be cleaned up by putting rxrpc_put_connection() out of line and making a tidy progression through the destruction code (offloaded to a workqueue if put from softirq or processor function context). The RCU part of the cleanup then only deals with the freeing at the end. (4) Make rxrpc_queue_conn() return immediately if it sees the active count is -1 rather then queuing the connection. (5) Make sure that the cleanup routine waits for the work item to complete. (6) Stash the rxrpc_net pointer in the conn struct so that the rcu free routine can use it, even if the local endpoint has been freed. Unfortunately, neither the timer nor the work item can simply get around the problem by just using refcount_inc_not_zero() as the waits would still have to be done, and there would still be the possibility of having to put the ref in the expiration function. Note the connection work item is mostly going to go away with the main event work being transferred to the I/O thread, so the wait in (6) will become obsolete. Signed-off-by: David Howells <dhowells@redhat.com> cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org
2022-11-25 12:43:50 +00:00
" SvID ConnID End Ref Act State Key "
rxrpc: Rewrite the client connection manager Rewrite the rxrpc client connection manager so that it can support multiple connections for a given security key to a peer. The following changes are made: (1) For each open socket, the code currently maintains an rbtree with the connections placed into it, keyed by communications parameters. This is tricky to maintain as connections can be culled from the tree or replaced within it. Connections can require replacement for a number of reasons, e.g. their IDs span too great a range for the IDR data type to represent efficiently, the call ID numbers on that conn would overflow or the conn got aborted. This is changed so that there's now a connection bundle object placed in the tree, keyed on the same parameters. The bundle, however, does not need to be replaced. (2) An rxrpc_bundle object can now manage the available channels for a set of parallel connections. The lock that manages this is moved there from the rxrpc_connection struct (channel_lock). (3) There'a a dummy bundle for all incoming connections to share so that they have a channel_lock too. It might be better to give each incoming connection its own bundle. This bundle is not needed to manage which channels incoming calls are made on because that's the solely at whim of the client. (4) The restrictions on how many client connections are around are removed. Instead, a previous patch limits the number of client calls that can be allocated. Ordinarily, client connections are reaped after 2 minutes on the idle queue, but when more than a certain number of connections are in existence, the reaper starts reaping them after 2s of idleness instead to get the numbers back down. It could also be made such that new call allocations are forced to wait until the number of outstanding connections subsides. Signed-off-by: David Howells <dhowells@redhat.com>
2020-07-01 10:15:32 +00:00
" Serial ISerial CallId0 CallId1 CallId2 CallId3\n"
);
return 0;
}
conn = list_entry(v, struct rxrpc_connection, proc_link);
rxrpc: Preallocate peers, conns and calls for incoming service requests Make it possible for the data_ready handler called from the UDP transport socket to completely instantiate an rxrpc_call structure and make it immediately live by preallocating all the memory it might need. The idea is to cut out the background thread usage as much as possible. [Note that the preallocated structs are not actually used in this patch - that will be done in a future patch.] If insufficient resources are available in the preallocation buffers, it will be possible to discard the DATA packet in the data_ready handler or schedule a BUSY packet without the need to schedule an attempt at allocation in a background thread. To this end: (1) Preallocate rxrpc_peer, rxrpc_connection and rxrpc_call structs to a maximum number each of the listen backlog size. The backlog size is limited to a maxmimum of 32. Only this many of each can be in the preallocation buffer. (2) For userspace sockets, the preallocation is charged initially by listen() and will be recharged by accepting or rejecting pending new incoming calls. (3) For kernel services {,re,dis}charging of the preallocation buffers is handled manually. Two notifier callbacks have to be provided before kernel_listen() is invoked: (a) An indication that a new call has been instantiated. This can be used to trigger background recharging. (b) An indication that a call is being discarded. This is used when the socket is being released. A function, rxrpc_kernel_charge_accept() is called by the kernel service to preallocate a single call. It should be passed the user ID to be used for that call and a callback to associate the rxrpc call with the kernel service's side of the ID. (4) Discard the preallocation when the socket is closed. (5) Temporarily bump the refcount on the call allocated in rxrpc_incoming_call() so that rxrpc_release_call() can ditch the preallocation ref on service calls unconditionally. This will no longer be necessary once the preallocation is used. Note that this does not yet control the number of active service calls on a client - that will come in a later patch. A future development would be to provide a setsockopt() call that allows a userspace server to manually charge the preallocation buffer. This would allow user call IDs to be provided in advance and the awkward manual accept stage to be bypassed. Signed-off-by: David Howells <dhowells@redhat.com>
2016-09-08 10:10:12 +00:00
if (conn->state == RXRPC_CONN_SERVICE_PREALLOC) {
strcpy(lbuff, "no_local");
strcpy(rbuff, "no_connection");
goto print;
}
sprintf(lbuff, "%pISpc", &conn->local->srx.transport);
sprintf(rbuff, "%pISpc", &conn->peer->srx.transport);
rxrpc: Preallocate peers, conns and calls for incoming service requests Make it possible for the data_ready handler called from the UDP transport socket to completely instantiate an rxrpc_call structure and make it immediately live by preallocating all the memory it might need. The idea is to cut out the background thread usage as much as possible. [Note that the preallocated structs are not actually used in this patch - that will be done in a future patch.] If insufficient resources are available in the preallocation buffers, it will be possible to discard the DATA packet in the data_ready handler or schedule a BUSY packet without the need to schedule an attempt at allocation in a background thread. To this end: (1) Preallocate rxrpc_peer, rxrpc_connection and rxrpc_call structs to a maximum number each of the listen backlog size. The backlog size is limited to a maxmimum of 32. Only this many of each can be in the preallocation buffer. (2) For userspace sockets, the preallocation is charged initially by listen() and will be recharged by accepting or rejecting pending new incoming calls. (3) For kernel services {,re,dis}charging of the preallocation buffers is handled manually. Two notifier callbacks have to be provided before kernel_listen() is invoked: (a) An indication that a new call has been instantiated. This can be used to trigger background recharging. (b) An indication that a call is being discarded. This is used when the socket is being released. A function, rxrpc_kernel_charge_accept() is called by the kernel service to preallocate a single call. It should be passed the user ID to be used for that call and a callback to associate the rxrpc call with the kernel service's side of the ID. (4) Discard the preallocation when the socket is closed. (5) Temporarily bump the refcount on the call allocated in rxrpc_incoming_call() so that rxrpc_release_call() can ditch the preallocation ref on service calls unconditionally. This will no longer be necessary once the preallocation is used. Note that this does not yet control the number of active service calls on a client - that will come in a later patch. A future development would be to provide a setsockopt() call that allows a userspace server to manually charge the preallocation buffer. This would allow user call IDs to be provided in advance and the awkward manual accept stage to be bypassed. Signed-off-by: David Howells <dhowells@redhat.com>
2016-09-08 10:10:12 +00:00
print:
state = rxrpc_is_conn_aborted(conn) ?
rxrpc_call_completions[conn->completion] :
rxrpc_conn_states[conn->state];
seq_printf(seq,
rxrpc: Don't hold a ref for connection workqueue Currently, rxrpc gives the connection's work item a ref on the connection when it queues it - and this is called from the timer expiration function. The problem comes when queue_work() fails (ie. the work item is already queued): the timer routine must put the ref - but this may cause the cleanup code to run. This has the unfortunate effect that the cleanup code may then be run in softirq context - which means that any spinlocks it might need to touch have to be guarded to disable softirqs (ie. they need a "_bh" suffix). (1) Don't give a ref to the work item. (2) Simplify handling of service connections by adding a separate active count so that the refcount isn't also used for this. (3) Connection destruction for both client and service connections can then be cleaned up by putting rxrpc_put_connection() out of line and making a tidy progression through the destruction code (offloaded to a workqueue if put from softirq or processor function context). The RCU part of the cleanup then only deals with the freeing at the end. (4) Make rxrpc_queue_conn() return immediately if it sees the active count is -1 rather then queuing the connection. (5) Make sure that the cleanup routine waits for the work item to complete. (6) Stash the rxrpc_net pointer in the conn struct so that the rcu free routine can use it, even if the local endpoint has been freed. Unfortunately, neither the timer nor the work item can simply get around the problem by just using refcount_inc_not_zero() as the waits would still have to be done, and there would still be the possibility of having to put the ref in the expiration function. Note the connection work item is mostly going to go away with the main event work being transferred to the I/O thread, so the wait in (6) will become obsolete. Signed-off-by: David Howells <dhowells@redhat.com> cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org
2022-11-25 12:43:50 +00:00
"UDP %-47.47s %-47.47s %4x %08x %s %3u %3d"
" %s %08x %08x %08x %08x %08x %08x %08x\n",
lbuff,
rbuff,
conn->service_id,
conn->proto.cid,
rxrpc_conn_is_service(conn) ? "Svc" : "Clt",
refcount_read(&conn->ref),
rxrpc: Don't hold a ref for connection workqueue Currently, rxrpc gives the connection's work item a ref on the connection when it queues it - and this is called from the timer expiration function. The problem comes when queue_work() fails (ie. the work item is already queued): the timer routine must put the ref - but this may cause the cleanup code to run. This has the unfortunate effect that the cleanup code may then be run in softirq context - which means that any spinlocks it might need to touch have to be guarded to disable softirqs (ie. they need a "_bh" suffix). (1) Don't give a ref to the work item. (2) Simplify handling of service connections by adding a separate active count so that the refcount isn't also used for this. (3) Connection destruction for both client and service connections can then be cleaned up by putting rxrpc_put_connection() out of line and making a tidy progression through the destruction code (offloaded to a workqueue if put from softirq or processor function context). The RCU part of the cleanup then only deals with the freeing at the end. (4) Make rxrpc_queue_conn() return immediately if it sees the active count is -1 rather then queuing the connection. (5) Make sure that the cleanup routine waits for the work item to complete. (6) Stash the rxrpc_net pointer in the conn struct so that the rcu free routine can use it, even if the local endpoint has been freed. Unfortunately, neither the timer nor the work item can simply get around the problem by just using refcount_inc_not_zero() as the waits would still have to be done, and there would still be the possibility of having to put the ref in the expiration function. Note the connection work item is mostly going to go away with the main event work being transferred to the I/O thread, so the wait in (6) will become obsolete. Signed-off-by: David Howells <dhowells@redhat.com> cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org
2022-11-25 12:43:50 +00:00
atomic_read(&conn->active),
state,
key_serial(conn->key),
conn->tx_serial,
conn->hi_serial,
conn->channels[0].call_id,
conn->channels[1].call_id,
conn->channels[2].call_id,
conn->channels[3].call_id);
return 0;
}
const struct seq_operations rxrpc_connection_seq_ops = {
.start = rxrpc_connection_seq_start,
.next = rxrpc_connection_seq_next,
.stop = rxrpc_connection_seq_stop,
.show = rxrpc_connection_seq_show,
};
/*
* generate a list of extant virtual bundles in /proc/net/rxrpc/bundles
*/
static void *rxrpc_bundle_seq_start(struct seq_file *seq, loff_t *_pos)
__acquires(rxnet->conn_lock)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
read_lock(&rxnet->conn_lock);
return seq_list_start_head(&rxnet->bundle_proc_list, *_pos);
}
static void *rxrpc_bundle_seq_next(struct seq_file *seq, void *v,
loff_t *pos)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
return seq_list_next(v, &rxnet->bundle_proc_list, pos);
}
static void rxrpc_bundle_seq_stop(struct seq_file *seq, void *v)
__releases(rxnet->conn_lock)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
read_unlock(&rxnet->conn_lock);
}
static int rxrpc_bundle_seq_show(struct seq_file *seq, void *v)
{
struct rxrpc_bundle *bundle;
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
char lbuff[50], rbuff[50];
if (v == &rxnet->bundle_proc_list) {
seq_puts(seq,
"Proto Local "
" Remote "
" SvID Ref Act Flg Key |"
" Bundle Conn_0 Conn_1 Conn_2 Conn_3\n"
);
return 0;
}
bundle = list_entry(v, struct rxrpc_bundle, proc_link);
sprintf(lbuff, "%pISpc", &bundle->local->srx.transport);
sprintf(rbuff, "%pISpc", &bundle->peer->srx.transport);
seq_printf(seq,
"UDP %-47.47s %-47.47s %4x %3u %3d"
" %c%c%c %08x | %08x %08x %08x %08x %08x\n",
lbuff,
rbuff,
bundle->service_id,
refcount_read(&bundle->ref),
atomic_read(&bundle->active),
bundle->try_upgrade ? 'U' : '-',
bundle->exclusive ? 'e' : '-',
bundle->upgrade ? 'u' : '-',
key_serial(bundle->key),
bundle->debug_id,
bundle->conn_ids[0],
bundle->conn_ids[1],
bundle->conn_ids[2],
bundle->conn_ids[3]);
return 0;
}
const struct seq_operations rxrpc_bundle_seq_ops = {
.start = rxrpc_bundle_seq_start,
.next = rxrpc_bundle_seq_next,
.stop = rxrpc_bundle_seq_stop,
.show = rxrpc_bundle_seq_show,
};
/*
* generate a list of extant virtual peers in /proc/net/rxrpc/peers
*/
static int rxrpc_peer_seq_show(struct seq_file *seq, void *v)
{
struct rxrpc_peer *peer;
time64_t now;
char lbuff[50], rbuff[50];
if (v == SEQ_START_TOKEN) {
seq_puts(seq,
"Proto Local "
" Remote "
" Use SST MTU LastUse RTT RTO\n"
);
return 0;
}
peer = list_entry(v, struct rxrpc_peer, hash_link);
sprintf(lbuff, "%pISpc", &peer->local->srx.transport);
sprintf(rbuff, "%pISpc", &peer->srx.transport);
now = ktime_get_seconds();
seq_printf(seq,
"UDP %-47.47s %-47.47s %3u"
rxrpc: Fix the excessive initial retransmission timeout rxrpc currently uses a fixed 4s retransmission timeout until the RTT is sufficiently sampled. This can cause problems with some fileservers with calls to the cache manager in the afs filesystem being dropped from the fileserver because a packet goes missing and the retransmission timeout is greater than the call expiry timeout. Fix this by: (1) Copying the RTT/RTO calculation code from Linux's TCP implementation and altering it to fit rxrpc. (2) Altering the various users of the RTT to make use of the new SRTT value. (3) Replacing the use of rxrpc_resend_timeout to use the calculated RTO value instead (which is needed in jiffies), along with a backoff. Notes: (1) rxrpc provides RTT samples by matching the serial numbers on outgoing DATA packets that have the RXRPC_REQUEST_ACK set and PING ACK packets against the reference serial number in incoming REQUESTED ACK and PING-RESPONSE ACK packets. (2) Each packet that is transmitted on an rxrpc connection gets a new per-connection serial number, even for retransmissions, so an ACK can be cross-referenced to a specific trigger packet. This allows RTT information to be drawn from retransmitted DATA packets also. (3) rxrpc maintains the RTT/RTO state on the rxrpc_peer record rather than on an rxrpc_call because many RPC calls won't live long enough to generate more than one sample. (4) The calculated SRTT value is in units of 8ths of a microsecond rather than nanoseconds. The (S)RTT and RTO values are displayed in /proc/net/rxrpc/peers. Fixes: 17926a79320a ([AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both"") Signed-off-by: David Howells <dhowells@redhat.com>
2020-05-11 13:54:34 +00:00
" %3u %5u %6llus %8u %8u\n",
lbuff,
rbuff,
refcount_read(&peer->ref),
peer->cong_ssthresh,
peer->mtu,
now - peer->last_tx_at,
rxrpc: Fix the excessive initial retransmission timeout rxrpc currently uses a fixed 4s retransmission timeout until the RTT is sufficiently sampled. This can cause problems with some fileservers with calls to the cache manager in the afs filesystem being dropped from the fileserver because a packet goes missing and the retransmission timeout is greater than the call expiry timeout. Fix this by: (1) Copying the RTT/RTO calculation code from Linux's TCP implementation and altering it to fit rxrpc. (2) Altering the various users of the RTT to make use of the new SRTT value. (3) Replacing the use of rxrpc_resend_timeout to use the calculated RTO value instead (which is needed in jiffies), along with a backoff. Notes: (1) rxrpc provides RTT samples by matching the serial numbers on outgoing DATA packets that have the RXRPC_REQUEST_ACK set and PING ACK packets against the reference serial number in incoming REQUESTED ACK and PING-RESPONSE ACK packets. (2) Each packet that is transmitted on an rxrpc connection gets a new per-connection serial number, even for retransmissions, so an ACK can be cross-referenced to a specific trigger packet. This allows RTT information to be drawn from retransmitted DATA packets also. (3) rxrpc maintains the RTT/RTO state on the rxrpc_peer record rather than on an rxrpc_call because many RPC calls won't live long enough to generate more than one sample. (4) The calculated SRTT value is in units of 8ths of a microsecond rather than nanoseconds. The (S)RTT and RTO values are displayed in /proc/net/rxrpc/peers. Fixes: 17926a79320a ([AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both"") Signed-off-by: David Howells <dhowells@redhat.com>
2020-05-11 13:54:34 +00:00
peer->srtt_us >> 3,
peer->rto_us);
return 0;
}
static void *rxrpc_peer_seq_start(struct seq_file *seq, loff_t *_pos)
__acquires(rcu)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
unsigned int bucket, n;
unsigned int shift = 32 - HASH_BITS(rxnet->peer_hash);
void *p;
rcu_read_lock();
if (*_pos >= UINT_MAX)
return NULL;
n = *_pos & ((1U << shift) - 1);
bucket = *_pos >> shift;
for (;;) {
if (bucket >= HASH_SIZE(rxnet->peer_hash)) {
*_pos = UINT_MAX;
return NULL;
}
if (n == 0) {
if (bucket == 0)
return SEQ_START_TOKEN;
*_pos += 1;
n++;
}
p = seq_hlist_start_rcu(&rxnet->peer_hash[bucket], n - 1);
if (p)
return p;
bucket++;
n = 1;
*_pos = (bucket << shift) | n;
}
}
static void *rxrpc_peer_seq_next(struct seq_file *seq, void *v, loff_t *_pos)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
unsigned int bucket, n;
unsigned int shift = 32 - HASH_BITS(rxnet->peer_hash);
void *p;
if (*_pos >= UINT_MAX)
return NULL;
bucket = *_pos >> shift;
p = seq_hlist_next_rcu(v, &rxnet->peer_hash[bucket], _pos);
if (p)
return p;
for (;;) {
bucket++;
n = 1;
*_pos = (bucket << shift) | n;
if (bucket >= HASH_SIZE(rxnet->peer_hash)) {
*_pos = UINT_MAX;
return NULL;
}
if (n == 0) {
*_pos += 1;
n++;
}
p = seq_hlist_start_rcu(&rxnet->peer_hash[bucket], n - 1);
if (p)
return p;
}
}
static void rxrpc_peer_seq_stop(struct seq_file *seq, void *v)
__releases(rcu)
{
rcu_read_unlock();
}
const struct seq_operations rxrpc_peer_seq_ops = {
.start = rxrpc_peer_seq_start,
.next = rxrpc_peer_seq_next,
.stop = rxrpc_peer_seq_stop,
.show = rxrpc_peer_seq_show,
};
/*
* Generate a list of extant virtual local endpoints in /proc/net/rxrpc/locals
*/
static int rxrpc_local_seq_show(struct seq_file *seq, void *v)
{
struct rxrpc_local *local;
char lbuff[50];
if (v == SEQ_START_TOKEN) {
seq_puts(seq,
"Proto Local "
" Use Act RxQ\n");
return 0;
}
local = hlist_entry(v, struct rxrpc_local, link);
sprintf(lbuff, "%pISpc", &local->srx.transport);
seq_printf(seq,
"UDP %-47.47s %3u %3u %3u\n",
lbuff,
refcount_read(&local->ref),
atomic_read(&local->active_users),
local->rx_queue.qlen);
return 0;
}
static void *rxrpc_local_seq_start(struct seq_file *seq, loff_t *_pos)
__acquires(rcu)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
unsigned int n;
rcu_read_lock();
if (*_pos >= UINT_MAX)
return NULL;
n = *_pos;
if (n == 0)
return SEQ_START_TOKEN;
return seq_hlist_start_rcu(&rxnet->local_endpoints, n - 1);
}
static void *rxrpc_local_seq_next(struct seq_file *seq, void *v, loff_t *_pos)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_net(seq));
if (*_pos >= UINT_MAX)
return NULL;
return seq_hlist_next_rcu(v, &rxnet->local_endpoints, _pos);
}
static void rxrpc_local_seq_stop(struct seq_file *seq, void *v)
__releases(rcu)
{
rcu_read_unlock();
}
const struct seq_operations rxrpc_local_seq_ops = {
.start = rxrpc_local_seq_start,
.next = rxrpc_local_seq_next,
.stop = rxrpc_local_seq_stop,
.show = rxrpc_local_seq_show,
};
/*
* Display stats in /proc/net/rxrpc/stats
*/
int rxrpc_stats_show(struct seq_file *seq, void *v)
{
struct rxrpc_net *rxnet = rxrpc_net(seq_file_single_net(seq));
seq_printf(seq,
"Data : send=%u sendf=%u fail=%u\n",
atomic_read(&rxnet->stat_tx_data_send),
atomic_read(&rxnet->stat_tx_data_send_frag),
atomic_read(&rxnet->stat_tx_data_send_fail));
seq_printf(seq,
"Data-Tx : nr=%u retrans=%u uf=%u cwr=%u\n",
atomic_read(&rxnet->stat_tx_data),
atomic_read(&rxnet->stat_tx_data_retrans),
atomic_read(&rxnet->stat_tx_data_underflow),
atomic_read(&rxnet->stat_tx_data_cwnd_reset));
seq_printf(seq,
"Data-Rx : nr=%u reqack=%u jumbo=%u\n",
atomic_read(&rxnet->stat_rx_data),
atomic_read(&rxnet->stat_rx_data_reqack),
atomic_read(&rxnet->stat_rx_data_jumbo));
seq_printf(seq,
"Ack : fill=%u send=%u skip=%u\n",
atomic_read(&rxnet->stat_tx_ack_fill),
atomic_read(&rxnet->stat_tx_ack_send),
atomic_read(&rxnet->stat_tx_ack_skip));
seq_printf(seq,
"Ack-Tx : req=%u dup=%u oos=%u exw=%u nos=%u png=%u prs=%u dly=%u idl=%u\n",
atomic_read(&rxnet->stat_tx_acks[RXRPC_ACK_REQUESTED]),
atomic_read(&rxnet->stat_tx_acks[RXRPC_ACK_DUPLICATE]),
atomic_read(&rxnet->stat_tx_acks[RXRPC_ACK_OUT_OF_SEQUENCE]),
atomic_read(&rxnet->stat_tx_acks[RXRPC_ACK_EXCEEDS_WINDOW]),
atomic_read(&rxnet->stat_tx_acks[RXRPC_ACK_NOSPACE]),
atomic_read(&rxnet->stat_tx_acks[RXRPC_ACK_PING]),
atomic_read(&rxnet->stat_tx_acks[RXRPC_ACK_PING_RESPONSE]),
atomic_read(&rxnet->stat_tx_acks[RXRPC_ACK_DELAY]),
atomic_read(&rxnet->stat_tx_acks[RXRPC_ACK_IDLE]));
seq_printf(seq,
"Ack-Rx : req=%u dup=%u oos=%u exw=%u nos=%u png=%u prs=%u dly=%u idl=%u\n",
atomic_read(&rxnet->stat_rx_acks[RXRPC_ACK_REQUESTED]),
atomic_read(&rxnet->stat_rx_acks[RXRPC_ACK_DUPLICATE]),
atomic_read(&rxnet->stat_rx_acks[RXRPC_ACK_OUT_OF_SEQUENCE]),
atomic_read(&rxnet->stat_rx_acks[RXRPC_ACK_EXCEEDS_WINDOW]),
atomic_read(&rxnet->stat_rx_acks[RXRPC_ACK_NOSPACE]),
atomic_read(&rxnet->stat_rx_acks[RXRPC_ACK_PING]),
atomic_read(&rxnet->stat_rx_acks[RXRPC_ACK_PING_RESPONSE]),
atomic_read(&rxnet->stat_rx_acks[RXRPC_ACK_DELAY]),
atomic_read(&rxnet->stat_rx_acks[RXRPC_ACK_IDLE]));
seq_printf(seq,
"Why-Req-A: acklost=%u already=%u mrtt=%u ortt=%u\n",
atomic_read(&rxnet->stat_why_req_ack[rxrpc_reqack_ack_lost]),
atomic_read(&rxnet->stat_why_req_ack[rxrpc_reqack_already_on]),
atomic_read(&rxnet->stat_why_req_ack[rxrpc_reqack_more_rtt]),
atomic_read(&rxnet->stat_why_req_ack[rxrpc_reqack_old_rtt]));
seq_printf(seq,
"Why-Req-A: nolast=%u retx=%u slows=%u smtxw=%u\n",
atomic_read(&rxnet->stat_why_req_ack[rxrpc_reqack_no_srv_last]),
atomic_read(&rxnet->stat_why_req_ack[rxrpc_reqack_retrans]),
atomic_read(&rxnet->stat_why_req_ack[rxrpc_reqack_slow_start]),
atomic_read(&rxnet->stat_why_req_ack[rxrpc_reqack_small_txwin]));
seq_printf(seq,
rxrpc: Don't use a ring buffer for call Tx queue Change the way the Tx queueing works to make the following ends easier to achieve: (1) The filling of packets, the encryption of packets and the transmission of packets can be handled in parallel by separate threads, rather than rxrpc_sendmsg() allocating, filling, encrypting and transmitting each packet before moving onto the next one. (2) Get rid of the fixed-size ring which sets a hard limit on the number of packets that can be retained in the ring. This allows the number of packets to increase without having to allocate a very large ring or having variable-sized rings. [Note: the downside of this is that it's then less efficient to locate a packet for retransmission as we then have to step through a list and examine each buffer in the list.] (3) Allow the filler/encrypter to run ahead of the transmission window. (4) Make it easier to do zero copy UDP from the packet buffers. (5) Make it easier to do zero copy from userspace to the packet buffers - and thence to UDP (only if for unauthenticated connections). To that end, the following changes are made: (1) Use the new rxrpc_txbuf struct instead of sk_buff for keeping packets to be transmitted in. This allows them to be placed on multiple queues simultaneously. An sk_buff isn't really necessary as it's never passed on to lower-level networking code. (2) Keep the transmissable packets in a linked list on the call struct rather than in a ring. As a consequence, the annotation buffer isn't used either; rather a flag is set on the packet to indicate ackedness. (3) Use the RXRPC_CALL_TX_LAST flag to indicate that the last packet to be transmitted has been queued. Add RXRPC_CALL_TX_ALL_ACKED to indicate that all packets up to and including the last got hard acked. (4) Wire headers are now stored in the txbuf rather than being concocted on the stack and they're stored immediately before the data, thereby allowing zerocopy of a single span. (5) Don't bother with instant-resend on transmission failure; rather, leave it for a timer or an ACK packet to trigger. Signed-off-by: David Howells <dhowells@redhat.com> cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org
2022-03-31 22:55:08 +00:00
"Buffers : txb=%u rxb=%u\n",
atomic_read(&rxrpc_nr_txbuf),
atomic_read(&rxrpc_n_rx_skbs));
seq_printf(seq,
"IO-thread: loops=%u\n",
atomic_read(&rxnet->stat_io_loop));
return 0;
}
/*
* Clear stats if /proc/net/rxrpc/stats is written to.
*/
int rxrpc_stats_clear(struct file *file, char *buf, size_t size)
{
struct seq_file *m = file->private_data;
struct rxrpc_net *rxnet = rxrpc_net(seq_file_single_net(m));
if (size > 1 || (size == 1 && buf[0] != '\n'))
return -EINVAL;
atomic_set(&rxnet->stat_tx_data, 0);
atomic_set(&rxnet->stat_tx_data_retrans, 0);
atomic_set(&rxnet->stat_tx_data_underflow, 0);
atomic_set(&rxnet->stat_tx_data_cwnd_reset, 0);
atomic_set(&rxnet->stat_tx_data_send, 0);
atomic_set(&rxnet->stat_tx_data_send_frag, 0);
atomic_set(&rxnet->stat_tx_data_send_fail, 0);
atomic_set(&rxnet->stat_rx_data, 0);
atomic_set(&rxnet->stat_rx_data_reqack, 0);
atomic_set(&rxnet->stat_rx_data_jumbo, 0);
atomic_set(&rxnet->stat_tx_ack_fill, 0);
atomic_set(&rxnet->stat_tx_ack_send, 0);
atomic_set(&rxnet->stat_tx_ack_skip, 0);
memset(&rxnet->stat_tx_acks, 0, sizeof(rxnet->stat_tx_acks));
memset(&rxnet->stat_rx_acks, 0, sizeof(rxnet->stat_rx_acks));
memset(&rxnet->stat_why_req_ack, 0, sizeof(rxnet->stat_why_req_ack));
atomic_set(&rxnet->stat_io_loop, 0);
return size;
}