[ Upstream commit 0c6de0c943dbb42831bf7502eb5c007f71e752d2 ]
FB_MTU is used in 'tipc_msg_build()' to alloc smaller skb when memory
allocation fails, which can avoid unnecessary sending failures.
The value of FB_MTU now is 3744, and the data size will be:
(3744 + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) + \
SKB_DATA_ALIGN(BUF_HEADROOM + BUF_TAILROOM + 3))
which is larger than one page(4096), and two pages will be allocated.
To avoid it, replace '3744' with a calculation:
(PAGE_SIZE - SKB_DATA_ALIGN(BUF_OVERHEAD) - \
SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
What's more, alloc_skb_fclone() will call SKB_DATA_ALIGN for data size,
and it's not necessary to make alignment for buf_size in
tipc_buf_acquire(). So, just remove it.
Fixes: 4c94cc2d3d ("tipc: fall back to smaller MTU if allocation of local send skb fails")
Signed-off-by: Menglong Dong <dong.menglong@zte.com.cn>
Acked-by: Jon Maloy <jmaloy@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit b7df21cf1b79ab7026f545e7bf837bd5750ac026 upstream.
It's not a good idea to append the frag skb to a skb's frag_list if
the frag_list already has skbs from elsewhere, such as this skb was
created by pskb_copy() where the frag_list was cloned (all the skbs
in it were skb_get'ed) and shared by multiple skbs.
However, the new appended frag skb should have been only seen by the
current skb. Otherwise, it will cause use after free crashes as this
appended frag skb are seen by multiple skbs but it only got skb_get
called once.
The same thing happens with a skb updated by pskb_may_pull() with a
skb_cloned skb. Li Shuang has reported quite a few crashes caused
by this when doing testing over macvlan devices:
[] kernel BUG at net/core/skbuff.c:1970!
[] Call Trace:
[] skb_clone+0x4d/0xb0
[] macvlan_broadcast+0xd8/0x160 [macvlan]
[] macvlan_process_broadcast+0x148/0x150 [macvlan]
[] process_one_work+0x1a7/0x360
[] worker_thread+0x30/0x390
[] kernel BUG at mm/usercopy.c:102!
[] Call Trace:
[] __check_heap_object+0xd3/0x100
[] __check_object_size+0xff/0x16b
[] simple_copy_to_iter+0x1c/0x30
[] __skb_datagram_iter+0x7d/0x310
[] __skb_datagram_iter+0x2a5/0x310
[] skb_copy_datagram_iter+0x3b/0x90
[] tipc_recvmsg+0x14a/0x3a0 [tipc]
[] ____sys_recvmsg+0x91/0x150
[] ___sys_recvmsg+0x7b/0xc0
[] kernel BUG at mm/slub.c:305!
[] Call Trace:
[] <IRQ>
[] kmem_cache_free+0x3ff/0x400
[] __netif_receive_skb_core+0x12c/0xc40
[] ? kmem_cache_alloc+0x12e/0x270
[] netif_receive_skb_internal+0x3d/0xb0
[] ? get_rx_page_info+0x8e/0xa0 [be2net]
[] be_poll+0x6ef/0xd00 [be2net]
[] ? irq_exit+0x4f/0x100
[] net_rx_action+0x149/0x3b0
...
This patch is to fix it by linearizing the head skb if it has frag_list
set in tipc_buf_append(). Note that we choose to do this before calling
skb_unshare(), as __skb_linearize() will avoid skb_copy(). Also, we can
not just drop the frag_list either as the early time.
Fixes: 45c8b7b175 ("tipc: allow non-linear first fragment buffer")
Reported-by: Li Shuang <shuali@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Jon Maloy <jmaloy@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit ed42989eab ("tipc: fix the skb_unshare() in tipc_buf_append()")
replaced skb_unshare() with skb_copy() to not reduce the data reference
counter of the original skb intentionally. This is not the correct
way to handle the cloned skb because it causes memory leak in 2
following cases:
1/ Sending multicast messages via broadcast link
The original skb list is cloned to the local skb list for local
destination. After that, the data reference counter of each skb
in the original list has the value of 2. This causes each skb not
to be freed after receiving ACK:
tipc_link_advance_transmq()
{
...
/* release skb */
__skb_unlink(skb, &l->transmq);
kfree_skb(skb); <-- memory exists after being freed
}
2/ Sending multicast messages via replicast link
Similar to the above case, each skb cannot be freed after purging
the skb list:
tipc_mcast_xmit()
{
...
__skb_queue_purge(pkts); <-- memory exists after being freed
}
This commit fixes this issue by using skb_unshare() instead. Besides,
to avoid use-after-free error reported by KASAN, the pointer to the
fragment is set to NULL before calling skb_unshare() to make sure that
the original skb is not freed after freeing the fragment 2 times in
case skb_unshare() returns NULL.
Fixes: ed42989eab ("tipc: fix the skb_unshare() in tipc_buf_append()")
Acked-by: Jon Maloy <jmaloy@redhat.com>
Reported-by: Thang Hoang Ngo <thang.h.ngo@dektech.com.au>
Signed-off-by: Tung Nguyen <tung.q.nguyen@dektech.com.au>
Reviewed-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Link: https://lore.kernel.org/r/20201027032403.1823-1-tung.q.nguyen@dektech.com.au
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Minor conflicts in net/mptcp/protocol.h and
tools/testing/selftests/net/Makefile.
In both cases code was added on both sides in the same place
so just keep both.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
skb_unshare() drops a reference count on the old skb unconditionally,
so in the failure case, we end up freeing the skb twice here.
And because the skb is allocated in fclone and cloned by caller
tipc_msg_reassemble(), the consequence is actually freeing the
original skb too, thus triggered the UAF by syzbot.
Fix this by replacing this skb_unshare() with skb_cloned()+skb_copy().
Fixes: ff48b6222e ("tipc: use skb_unshare() instead in tipc_buf_append()")
Reported-and-tested-by: syzbot+e96a7ba46281824cc46a@syzkaller.appspotmail.com
Cc: Jon Maloy <jmaloy@redhat.com>
Cc: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Two minor conflicts:
1) net/ipv4/route.c, adding a new local variable while
moving another local variable and removing it's
initial assignment.
2) drivers/net/dsa/microchip/ksz9477.c, overlapping changes.
One pretty prints the port mode differently, whilst another
changes the driver to try and obtain the port mode from
the port node rather than the switch node.
Signed-off-by: David S. Miller <davem@davemloft.net>
Drop repeated words in net/tipc/.
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Jon Maloy <jmaloy@redhat.com>
Cc: Ying Xue <ying.xue@windriver.com>
Cc: tipc-discussion@lists.sourceforge.net
Signed-off-by: David S. Miller <davem@davemloft.net>
In tipc_buf_append() it may change skb's frag_list, and it causes
problems when this skb is cloned. skb_unclone() doesn't really
make this skb's flag_list available to change.
Shuang Li has reported an use-after-free issue because of this
when creating quite a few macvlan dev over the same dev, where
the broadcast packets will be cloned and go up to the stack:
[ ] BUG: KASAN: use-after-free in pskb_expand_head+0x86d/0xea0
[ ] Call Trace:
[ ] dump_stack+0x7c/0xb0
[ ] print_address_description.constprop.7+0x1a/0x220
[ ] kasan_report.cold.10+0x37/0x7c
[ ] check_memory_region+0x183/0x1e0
[ ] pskb_expand_head+0x86d/0xea0
[ ] process_backlog+0x1df/0x660
[ ] net_rx_action+0x3b4/0xc90
[ ]
[ ] Allocated by task 1786:
[ ] kmem_cache_alloc+0xbf/0x220
[ ] skb_clone+0x10a/0x300
[ ] macvlan_broadcast+0x2f6/0x590 [macvlan]
[ ] macvlan_process_broadcast+0x37c/0x516 [macvlan]
[ ] process_one_work+0x66a/0x1060
[ ] worker_thread+0x87/0xb10
[ ]
[ ] Freed by task 3253:
[ ] kmem_cache_free+0x82/0x2a0
[ ] skb_release_data+0x2c3/0x6e0
[ ] kfree_skb+0x78/0x1d0
[ ] tipc_recvmsg+0x3be/0xa40 [tipc]
So fix it by using skb_unshare() instead, which would create a new
skb for the cloned frag and it'll be safe to change its frag_list.
The similar things were also done in sctp_make_reassembled_event(),
which is using skb_copy().
Reported-by: Shuang Li <shuali@redhat.com>
Fixes: 37e22164a8 ("tipc: rename and move message reassembly function")
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Simple fixes which require no deep knowledge of the code.
Cc: Jon Maloy <jmaloy@redhat.com>
Cc: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Jon Maloy <jmaloy@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
syzbot found the following issue:
WARNING: CPU: 0 PID: 6808 at include/linux/thread_info.h:150 check_copy_size include/linux/thread_info.h:150 [inline]
WARNING: CPU: 0 PID: 6808 at include/linux/thread_info.h:150 copy_from_iter include/linux/uio.h:144 [inline]
WARNING: CPU: 0 PID: 6808 at include/linux/thread_info.h:150 tipc_msg_append+0x49a/0x5e0 net/tipc/msg.c:242
Kernel panic - not syncing: panic_on_warn set ...
This happens after commit 5e9eeccc58 ("tipc: fix NULL pointer
dereference in streaming") that tried to build at least one buffer even
when the message data length is zero... However, it now exposes another
bug that the 'mss' can be zero and the 'cpy' will be negative, thus the
above kernel WARNING will appear!
The zero value of 'mss' is never expected because it means Nagle is not
enabled for the socket (actually the socket type was 'SOCK_SEQPACKET'),
so the function 'tipc_msg_append()' must not be called at all. But that
was in this particular case since the message data length was zero, and
the 'send <= maxnagle' check became true.
We resolve the issue by explicitly checking if Nagle is enabled for the
socket, i.e. 'maxnagle != 0' before calling the 'tipc_msg_append()'. We
also reinforce the function to against such a negative values if any.
Reported-by: syzbot+75139a7d2605236b0b7f@syzkaller.appspotmail.com
Fixes: c0bceb97db ("tipc: add smart nagle feature")
Acked-by: Jon Maloy <jmaloy@redhat.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
syzbot found the following crash:
general protection fault, probably for non-canonical address 0xdffffc0000000019: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x00000000000000c8-0x00000000000000cf]
CPU: 1 PID: 7060 Comm: syz-executor394 Not tainted 5.7.0-rc6-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:__tipc_sendstream+0xbde/0x11f0 net/tipc/socket.c:1591
Code: 00 00 00 00 48 39 5c 24 28 48 0f 44 d8 e8 fa 3e db f9 48 b8 00 00 00 00 00 fc ff df 48 8d bb c8 00 00 00 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 e2 04 00 00 48 8b 9b c8 00 00 00 48 b8 00 00 00
RSP: 0018:ffffc90003ef7818 EFLAGS: 00010202
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffffff8797fd9d
RDX: 0000000000000019 RSI: ffffffff8797fde6 RDI: 00000000000000c8
RBP: ffff888099848040 R08: ffff88809a5f6440 R09: fffffbfff1860b4c
R10: ffffffff8c305a5f R11: fffffbfff1860b4b R12: ffff88809984857e
R13: 0000000000000000 R14: ffff888086aa4000 R15: 0000000000000000
FS: 00000000009b4880(0000) GS:ffff8880ae700000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000020000140 CR3: 00000000a7fdf000 CR4: 00000000001406e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
tipc_sendstream+0x4c/0x70 net/tipc/socket.c:1533
sock_sendmsg_nosec net/socket.c:652 [inline]
sock_sendmsg+0xcf/0x120 net/socket.c:672
____sys_sendmsg+0x32f/0x810 net/socket.c:2352
___sys_sendmsg+0x100/0x170 net/socket.c:2406
__sys_sendmmsg+0x195/0x480 net/socket.c:2496
__do_sys_sendmmsg net/socket.c:2525 [inline]
__se_sys_sendmmsg net/socket.c:2522 [inline]
__x64_sys_sendmmsg+0x99/0x100 net/socket.c:2522
do_syscall_64+0xf6/0x7d0 arch/x86/entry/common.c:295
entry_SYSCALL_64_after_hwframe+0x49/0xb3
RIP: 0033:0x440199
...
This bug was bisected to commit 0a3e060f34 ("tipc: add test for Nagle
algorithm effectiveness"). However, it is not the case, the trouble was
from the base in the case of zero data length message sending, we would
unexpectedly make an empty 'txq' queue after the 'tipc_msg_append()' in
Nagle mode.
A similar crash can be generated even without the bisected patch but at
the link layer when it accesses the empty queue.
We solve the issues by building at least one buffer to go with socket's
header and an optional data section that may be empty like what we had
with the 'tipc_msg_build()'.
Note: the previous commit 4c21daae3d ("tipc: Fix NULL pointer
dereference in __tipc_sendstream()") is obsoleted by this one since the
'txq' will be never empty and the check of 'skb != NULL' is unnecessary
but it is safe anyway.
Reported-by: syzbot+8eac6d030e7807c21d32@syzkaller.appspotmail.com
Fixes: c0bceb97db ("tipc: add smart nagle feature")
Acked-by: Jon Maloy <jmaloy@redhat.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes gcc '-Wunused-but-set-variable' warning:
net/tipc/msg.c: In function 'tipc_msg_append':
net/tipc/msg.c:215:24: warning:
variable 'prev' set but not used [-Wunused-but-set-variable]
commit 0a3e060f34 ("tipc: add test for Nagle algorithm effectiveness")
left behind this, remove it.
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When streaming in Nagle mode, we try to bundle small messages from user
as many as possible if there is one outstanding buffer, i.e. not ACK-ed
by the receiving side, which helps boost up the overall throughput. So,
the algorithm's effectiveness really depends on when Nagle ACK comes or
what the specific network latency (RTT) is, compared to the user's
message sending rate.
In a bad case, the user's sending rate is low or the network latency is
small, there will not be many bundles, so making a Nagle ACK or waiting
for it is not meaningful.
For example: a user sends its messages every 100ms and the RTT is 50ms,
then for each messages, we require one Nagle ACK but then there is only
one user message sent without any bundles.
In a better case, even if we have a few bundles (e.g. the RTT = 300ms),
but now the user sends messages in medium size, then there will not be
any difference at all, that says 3 x 1000-byte data messages if bundled
will still result in 3 bundles with MTU = 1500.
When Nagle is ineffective, the delay in user message sending is clearly
wasted instead of sending directly.
Besides, adding Nagle ACKs will consume some processor load on both the
sending and receiving sides.
This commit adds a test on the effectiveness of the Nagle algorithm for
an individual connection in the network on which it actually runs.
Particularly, upon receipt of a Nagle ACK we will compare the number of
bundles in the backlog queue to the number of user messages which would
be sent directly without Nagle. If the ratio is good (e.g. >= 2), Nagle
mode will be kept for further message sending. Otherwise, we will leave
Nagle and put a 'penalty' on the connection, so it will have to spend
more 'one-way' messages before being able to re-enter Nagle.
In addition, the 'ack-required' bit is only set when really needed that
the number of Nagle ACKs will be reduced during Nagle mode.
Testing with benchmark showed that with the patch, there was not much
difference in throughput for small messages since the tool continuously
sends messages without a break, so Nagle would still take in effect.
Acked-by: Ying Xue <ying.xue@windriver.com>
Acked-by: Jon Maloy <jmaloy@redhat.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit enables dumping the statistics of a broadcast-receiver link
like the traditional 'broadcast-link' one (which is for broadcast-
sender). The link dumping can be triggered via netlink (e.g. the
iproute2/tipc tool) by the link flag - 'TIPC_NLA_LINK_BROADCAST' as the
indicator.
The name of a broadcast-receiver link of a specific peer will be in the
format: 'broadcast-link:<peer-id>'.
For example:
Link <broadcast-link:1001002>
Window:50 packets
RX packets:7841 fragments:2408/440 bundles:0/0
TX packets:0 fragments:0/0 bundles:0/0
RX naks:0 defs:124 dups:0
TX naks:21 acks:0 retrans:0
Congestion link:0 Send queue max:0 avg:0
In addition, the broadcast-receiver link statistics can be reset in the
usual way via netlink by specifying that link name in command.
Note: the 'tipc_link_name_ext()' is removed because the link name can
now be retrieved simply via the 'l->name'.
Acked-by: Ying Xue <ying.xue@windriver.com>
Acked-by: Jon Maloy <jmaloy@redhat.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Checking and returning 'true' boolean is useless as it will be
returning at end of function
Signed-off-by: Hoang Le <hoang.h.le@dektech.com.au>
Acked-by: Ying Xue <ying.xue@windriver.com>
Acked-by: Jon Maloy <jmaloy@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit offers an option to encrypt and authenticate all messaging,
including the neighbor discovery messages. The currently most advanced
algorithm supported is the AEAD AES-GCM (like IPSec or TLS). All
encryption/decryption is done at the bearer layer, just before leaving
or after entering TIPC.
Supported features:
- Encryption & authentication of all TIPC messages (header + data);
- Two symmetric-key modes: Cluster and Per-node;
- Automatic key switching;
- Key-expired revoking (sequence number wrapped);
- Lock-free encryption/decryption (RCU);
- Asynchronous crypto, Intel AES-NI supported;
- Multiple cipher transforms;
- Logs & statistics;
Two key modes:
- Cluster key mode: One single key is used for both TX & RX in all
nodes in the cluster.
- Per-node key mode: Each nodes in the cluster has one specific TX key.
For RX, a node requires its peers' TX key to be able to decrypt the
messages from those peers.
Key setting from user-space is performed via netlink by a user program
(e.g. the iproute2 'tipc' tool).
Internal key state machine:
Attach Align(RX)
+-+ +-+
| V | V
+---------+ Attach +---------+
| IDLE |---------------->| PENDING |(user = 0)
+---------+ +---------+
A A Switch| A
| | | |
| | Free(switch/revoked) | |
(Free)| +----------------------+ | |Timeout
| (TX) | | |(RX)
| | | |
| | v |
+---------+ Switch +---------+
| PASSIVE |<----------------| ACTIVE |
+---------+ (RX) +---------+
(user = 1) (user >= 1)
The number of TFMs is 10 by default and can be changed via the procfs
'net/tipc/max_tfms'. At this moment, as for simplicity, this file is
also used to print the crypto statistics at runtime:
echo 0xfff1 > /proc/sys/net/tipc/max_tfms
The patch defines a new TIPC version (v7) for the encryption message (-
backward compatibility as well). The message is basically encapsulated
as follows:
+----------------------------------------------------------+
| TIPCv7 encryption | Original TIPCv2 | Authentication |
| header | packet (encrypted) | Tag |
+----------------------------------------------------------+
The throughput is about ~40% for small messages (compared with non-
encryption) and ~9% for large messages. With the support from hardware
crypto i.e. the Intel AES-NI CPU instructions, the throughput increases
upto ~85% for small messages and ~55% for large messages.
By default, the new feature is inactive (i.e. no encryption) until user
sets a key for TIPC. There is however also a new option - "TIPC_CRYPTO"
in the kernel configuration to enable/disable the new code when needed.
MAINTAINERS | add two new files 'crypto.h' & 'crypto.c' in tipc
Acked-by: Ying Xue <ying.xue@windreiver.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
As mentioned in commit e95584a889 ("tipc: fix unlimited bundling of
small messages"), the current message bundling algorithm is inefficient
that can generate bundles of only one payload message, that causes
unnecessary overheads for both the sender and receiver.
This commit re-designs the 'tipc_msg_make_bundle()' function (now named
as 'tipc_msg_try_bundle()'), so that when a message comes at the first
place, we will just check & keep a reference to it if the message is
suitable for bundling. The message buffer will be put into the link
backlog queue and processed as normal. Later on, when another one comes
we will make a bundle with the first message if possible and so on...
This way, a bundle if really needed will always consist of at least two
payload messages. Otherwise, we let the first buffer go its way without
any need of bundling, so reduce the overheads to zero.
Moreover, since now we have both the messages in hand, we can even
optimize the 'tipc_msg_bundle()' function, make bundle of a very large
(size ~ MSS) and small messages which is not with the current algorithm
e.g. [1400-byte message] + [10-byte message] (MTU = 1500).
Acked-by: Ying Xue <ying.xue@windreiver.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
We introduce a feature that works like a combination of TCP_NAGLE and
TCP_CORK, but without some of the weaknesses of those. In particular,
we will not observe long delivery delays because of delayed acks, since
the algorithm itself decides if and when acks are to be sent from the
receiving peer.
- The nagle property as such is determined by manipulating a new
'maxnagle' field in struct tipc_sock. If certain conditions are met,
'maxnagle' will define max size of the messages which can be bundled.
If it is set to zero no messages are ever bundled, implying that the
nagle property is disabled.
- A socket with the nagle property enabled enters nagle mode when more
than 4 messages have been sent out without receiving any data message
from the peer.
- A socket leaves nagle mode whenever it receives a data message from
the peer.
In nagle mode, messages smaller than 'maxnagle' are accumulated in the
socket write queue. The last buffer in the queue is marked with a new
'ack_required' bit, which forces the receiving peer to send a CONN_ACK
message back to the sender upon reception.
The accumulated contents of the write queue is transmitted when one of
the following events or conditions occur.
- A CONN_ACK message is received from the peer.
- A data message is received from the peer.
- A SOCK_WAKEUP pseudo message is received from the link level.
- The write queue contains more than 64 1k blocks of data.
- The connection is being shut down.
- There is no CONN_ACK message to expect. I.e., there is currently
no outstanding message where the 'ack_required' bit was set. As a
consequence, the first message added after we enter nagle mode
is always sent directly with this bit set.
This new feature gives a 50-100% improvement of throughput for small
(i.e., less than MTU size) messages, while it might add up to one RTT
to latency time when the socket is in nagle mode.
Acked-by: Ying Xue <ying.xue@windreiver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We have identified a problem with the "oversubscription" policy in the
link transmission code.
When small messages are transmitted, and the sending link has reached
the transmit window limit, those messages will be bundled and put into
the link backlog queue. However, bundles of data messages are counted
at the 'CRITICAL' level, so that the counter for that level, instead of
the counter for the real, bundled message's level is the one being
increased.
Subsequent, to-be-bundled data messages at non-CRITICAL levels continue
to be tested against the unchanged counter for their own level, while
contributing to an unrestrained increase at the CRITICAL backlog level.
This leaves a gap in congestion control algorithm for small messages
that can result in starvation for other users or a "real" CRITICAL
user. Even that eventually can lead to buffer exhaustion & link reset.
We fix this by keeping a 'target_bskb' buffer pointer at each levels,
then when bundling, we only bundle messages at the same importance
level only. This way, we know exactly how many slots a certain level
have occupied in the queue, so can manage level congestion accurately.
By bundling messages at the same level, we even have more benefits. Let
consider this:
- One socket sends 64-byte messages at the 'CRITICAL' level;
- Another sends 4096-byte messages at the 'LOW' level;
When a 64-byte message comes and is bundled the first time, we put the
overhead of message bundle to it (+ 40-byte header, data copy, etc.)
for later use, but the next message can be a 4096-byte one that cannot
be bundled to the previous one. This means the last bundle carries only
one payload message which is totally inefficient, as for the receiver
also! Later on, another 64-byte message comes, now we make a new bundle
and the same story repeats...
With the new bundling algorithm, this will not happen, the 64-byte
messages will be bundled together even when the 4096-byte message(s)
comes in between. However, if the 4096-byte messages are sent at the
same level i.e. 'CRITICAL', the bundling algorithm will again cause the
same overhead.
Also, the same will happen even with only one socket sending small
messages at a rate close to the link transmit's one, so that, when one
message is bundled, it's transmitted shortly. Then, another message
comes, a new bundle is created and so on...
We will solve this issue radically by another patch.
Fixes: 365ad353c2 ("tipc: reduce risk of user starvation during link congestion")
Reported-by: Hoang Le <hoang.h.le@dektech.com.au>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
In conjunction with changing the interfaces' MTU (e.g. especially in
the case of a bonding) where the TIPC links are brought up and down
in a short time, a couple of issues were detected with the current link
changeover mechanism:
1) When one link is up but immediately forced down again, the failover
procedure will be carried out in order to failover all the messages in
the link's transmq queue onto the other working link. The link and node
state is also set to FAILINGOVER as part of the process. The message
will be transmited in form of a FAILOVER_MSG, so its size is plus of 40
bytes (= the message header size). There is no problem if the original
message size is not larger than the link's MTU - 40, and indeed this is
the max size of a normal payload messages. However, in the situation
above, because the link has just been up, the messages in the link's
transmq are almost SYNCH_MSGs which had been generated by the link
synching procedure, then their size might reach the max value already!
When the FAILOVER_MSG is built on the top of such a SYNCH_MSG, its size
will exceed the link's MTU. As a result, the messages are dropped
silently and the failover procedure will never end up, the link will
not be able to exit the FAILINGOVER state, so cannot be re-established.
2) The same scenario above can happen more easily in case the MTU of
the links is set differently or when changing. In that case, as long as
a large message in the failure link's transmq queue was built and
fragmented with its link's MTU > the other link's one, the issue will
happen (there is no need of a link synching in advance).
3) The link synching procedure also faces with the same issue but since
the link synching is only started upon receipt of a SYNCH_MSG, dropping
the message will not result in a state deadlock, but it is not expected
as design.
The 1) & 3) issues are resolved by the last commit that only a dummy
SYNCH_MSG (i.e. without data) is generated at the link synching, so the
size of a FAILOVER_MSG if any then will never exceed the link's MTU.
For the 2) issue, the only solution is trying to fragment the messages
in the failure link's transmq queue according to the working link's MTU
so they can be failovered then. A new function is made to accomplish
this, it will still be a TUNNEL PROTOCOL/FAILOVER MSG but if the
original message size is too large, it will be fragmented & reassembled
at the receiving side.
Acked-by: Ying Xue <ying.xue@windriver.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Default socket receive buffer size for a listener socket is 2Mb. For
each arriving empty SYN, the linux kernel allocates a 768 bytes buffer.
This means that a listener socket can serve maximum 2700 simultaneous
empty connection setup requests before it hits a receive buffer
overflow, and much fewer if the SYN is carrying any significant
amount of data.
When this happens the setup request is rejected, and the client
receives an ECONNREFUSED error.
This commit mitigates this problem by letting the client socket try to
retransmit the SYN message multiple times when it sees it rejected with
the code TIPC_ERR_OVERLOAD. Retransmission is done at random intervals
in the range of [100 ms, setup_timeout / 4], as many times as there is
room for within the setup timeout limit.
Signed-off-by: Tung Nguyen <tung.q.nguyen@dektech.com.au>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The function tipc_msg_reverse() is reversing the header of a message
while reusing the original buffer. We have seen at several occasions
that this may have unfortunate side effects when the buffer to be
reversed is a clone.
In one of the following commits we will again need to reverse cloned
buffers, so this is the right time to permanently eliminate this
problem. In this commit we let the said function always consume the
original buffer and replace it with a new one when applicable.
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The function tipc_msg_extract() is using skb_clone() to clone inner
messages from a message bundle buffer. Although this method is safe,
it has an undesired effect that each buffer clone inherits the
true-size of the bundling buffer. As a result, the buffer clone
almost always ends up with being copied anyway by the message
validation function. This makes the cloning into a sub-optimization.
In this commit we take the consequence of this realization, and copy
each inner message to a separately allocated buffer up front in the
extraction function.
As a bonus we can now eliminate the two cases where we had to copy
re-routed packets that may potentially go out on the wire again.
Signed-off-by: Tung Nguyen <tung.q.nguyen@dektech.com.au>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Publications for TIPC_CLUSTER_SCOPE and TIPC_ZONE_SCOPE are in all
aspects handled the same way, both on the publishing node and on the
receiving nodes.
Despite previous ambitions to the contrary, this is never going to change,
so we take the conseqeunce of this and obsolete TIPC_ZONE_SCOPE and related
macros/functions. Whenever a user is doing a bind() or a sendmsg() attempt
using ZONE_SCOPE we translate this internally to CLUSTER_SCOPE, while we
remain compatible with users and remote nodes still using ZONE_SCOPE.
Furthermore, the non-formalized scope value 0 has always been permitted
for use during lookup, with the same meaning as ZONE_SCOPE/CLUSTER_SCOPE.
We now permit it even as binding scope, but for compatibility reasons we
choose to not change the value of TIPC_CLUSTER_SCOPE.
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit d618d09a68 ("tipc: enforce valid ratio between skb truesize
and contents") we introduced a test for ensuring that the condition
truesize/datasize <= 4 is true for a received buffer. Unfortunately this
test has two problems.
- Because of the integer arithmetics the test
if (skb->truesize / buf_roundup_len(skb) > 4) will miss all
ratios [4 < ratio < 5], which was not the intention.
- The buffer returned by skb_copy() inherits skb->truesize of the
original buffer, which doesn't help the situation at all.
In this commit, we change the ratio condition and replace skb_copy()
with a call to skb_copy_expand() to finally get this right.
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When sending node local messages the code is using an 'mtu' of 66060
bytes to avoid unnecessary fragmentation. During situations of low
memory tipc_msg_build() may sometimes fail to allocate such large
buffers, resulting in unnecessary send failures. This can easily be
remedied by falling back to a smaller MTU, and then reassemble the
buffer chain as if the message were arriving from a remote node.
At the same time, we change the initial MTU setting of the broadcast
link to a lower value, so that large messages always are fragmented
into smaller buffers even when we run in single node mode. Apart from
obtaining the same advantage as for the 'fallback' solution above, this
turns out to give a significant performance improvement. This can
probably be explained with the __pskb_copy() operation performed on the
buffer for each recipient during reception. We found the optimal value
for this, considering the most relevant skb pool, to be 3744 bytes.
Acked-by: Ying Xue <ying.xue@ericsson.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The socket level flow control is based on the assumption that incoming
buffers meet the condition (skb->truesize / roundup(skb->len) <= 4),
where the latter value is rounded off upwards to the nearest 1k number.
This does empirically hold true for the device drivers we know, but we
cannot trust that it will always be so, e.g., in a system with jumbo
frames and very small packets.
We now introduce a check for this condition at packet arrival, and if
we find it to be false, we copy the packet to a new, smaller buffer,
where the condition will be true. We expect this to affect only a small
fraction of all incoming packets, if at all.
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the following commits we will need to handle multiple incoming and
rejected/returned buffers in the function socket.c::filter_rcv().
As a preparation for this, we generalize the function by handling
buffer queues instead of individual buffers. We also introduce a
help function tipc_skb_reject(), and rename filter_rcv() to
tipc_sk_filter_rcv() in line with other functions in socket.c.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a bundling message is received, the function tipc_link_input()
calls function tipc_msg_extract() to unbundle all inner messages of
the bundling message before adding them to input queue.
The function tipc_msg_extract() just clones all inner skb for all
inner messagges from the bundling skb. This means that the skb
headroom of an inner message overlaps with the data part of the
preceding message in the bundle.
If the message in question is a name addressed message, it may be
subject to a secondary destination lookup, and eventually be sent out
on one of the interfaces again. But, since what is perceived as headroom
by the device driver in reality is the last bytes of the preceding
message in the bundle, the latter will be overwritten by the MAC
addresses of the L2 header. If the preceding message has not yet been
consumed by the user, it will evenually be delivered with corrupted
contents.
This commit fixes this by uncloning all messages passing through the
function tipc_msg_lookup_dest(), hence ensuring that the headroom
is always valid when the message is passed on.
Signed-off-by: Tung Nguyen <tung.q.nguyen@dektech.com.au>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit e3a77561e7 ("tipc: split up function tipc_msg_eval()"),
we have updated the function tipc_msg_lookup_dest() to set the error
codes to negative values at destination lookup failures. Thus when
the function sets the error code to -TIPC_ERR_NO_NAME, its inserted
into the 4 bit error field of the message header as 0xf instead of
TIPC_ERR_NO_NAME (1). The value 0xf is an unknown error code.
In this commit, we set only positive error code.
Fixes: e3a77561e7 ("tipc: split up function tipc_msg_eval()")
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In tipc_msg_reverse(), we assign skb attributes to local pointers
in stack at startup. This is followed by skb_linearize() and for
cloned buffers we perform skb relocation using pskb_expand_head().
Both these methods may update the skb attributes and thus making
the pointers incorrect.
In this commit, we fix this error by ensuring that the pointers
are re-assigned after any of these skb operations.
Fixes: 29042e19f2 ("tipc: let function tipc_msg_reverse() expand header
when needed")
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the function msg_reverse(), we reverse the header while trying to
reuse the original buffer whenever possible. Those rejected/returned
messages are always transmitted as unicast, but the msg_non_seq field
is not explicitly set to zero as it should be.
We have seen cases where multicast senders set the message type to
"NOT dest_droppable", meaning that a multicast message shorter than
one MTU will be returned, e.g., during receive buffer overflow, by
reusing the original buffer. This has the effect that even the
'msg_non_seq' field is inadvertently inherited by the rejected message,
although it is now sent as a unicast message. This again leads the
receiving unicast link endpoint to steer the packet toward the broadcast
link receive function, where it is dropped. The affected unicast link is
thereafter (after 100 failed retransmissions) declared 'stale' and
reset.
We fix this by unconditionally setting the 'msg_non_seq' flag to zero
for all rejected/returned messages.
Reported-by: Canh Duc Luu <canh.d.luu@dektech.com.au>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The kernel may sleep under a rcu read lock in tipc_msg_reverse, and the
function call path is:
tipc_l2_rcv_msg (acquire the lock by rcu_read_lock)
tipc_rcv
tipc_sk_rcv
tipc_msg_reverse
pskb_expand_head(GFP_KERNEL) --> may sleep
tipc_node_broadcast
tipc_node_xmit_skb
tipc_node_xmit
tipc_sk_rcv
tipc_msg_reverse
pskb_expand_head(GFP_KERNEL) --> may sleep
To fix it, "GFP_KERNEL" is replaced with "GFP_ATOMIC".
Signed-off-by: Jia-Ju Bai <baijiaju1990@163.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TIPC multicast messages are currently carried over a reliable
'broadcast link', making use of the underlying media's ability to
transport packets as L2 broadcast or IP multicast to all nodes in
the cluster.
When the used bearer is lacking that ability, we can instead emulate
the broadcast service by replicating and sending the packets over as
many unicast links as needed to reach all identified destinations.
We now introduce a new TIPC link-level 'replicast' service that does
this.
Reviewed-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Until now, we allocate memory always with GFP_ATOMIC flag.
When the system is under memory pressure and a user tries to send,
the send fails due to low memory. However, the user application
can wait for free memory if we allocate it using GFP_KERNEL flag.
In this commit, we use allocate memory with GFP_KERNEL for all user
allocation.
Reported-by: Rune Torgersen <runet@innovsys.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
copy_from_iter_full(), copy_from_iter_full_nocache() and
csum_and_copy_from_iter_full() - counterparts of copy_from_iter()
et.al., advancing iterator only in case of successful full copy
and returning whether it had been successful or not.
Convert some obvious users. *NOTE* - do not blindly assume that
something is a good candidate for those unless you are sure that
not advancing iov_iter in failure case is the right thing in
this case. Anything that does short read/short write kind of
stuff (or is in a loop, etc.) is unlikely to be a good one.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
When extracting an individual message from a received "bundle" buffer,
we just create a clone of the base buffer, and adjust it to point into
the right position of the linearized data area of the latter. This works
well for regular message reception, but during periods of extremely high
load it may happen that an extracted buffer, e.g, a connection probe, is
reversed and forwarded through an external interface while the preceding
extracted message is still unhandled. When this happens, the header or
data area of the preceding message will be partially overwritten by a
MAC header, leading to unpredicatable consequences, such as a link
reset.
We now fix this by ensuring that the msg_reverse() function never
returns a cloned buffer, and that the returned buffer always contains
sufficient valid head and tail room to be forwarded.
Reported-by: Erik Hugne <erik.hugne@gmail.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The code path for receiving broadcast packets is currently distinct
from the unicast path. This leads to unnecessary code and data
duplication, something that can be avoided with some effort.
We now introduce separate per-peer tipc_link instances for handling
broadcast packet reception. Each receive link keeps a pointer to the
common, single, broadcast link instance, and can hence handle release
and retransmission of send buffers as if they belonged to the own
instance.
Furthermore, we let each unicast link instance keep a reference to both
the pertaining broadcast receive link, and to the common send link.
This makes it possible for the unicast links to easily access data for
broadcast link synchronization, as well as for carrying acknowledges for
received broadcast packets.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit simplifies the broadcast link transmission function, by
leveraging previous changes to the link transmission function and the
broadcast transmission link life cycle.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
net/ipv6/xfrm6_output.c
net/openvswitch/flow_netlink.c
net/openvswitch/vport-gre.c
net/openvswitch/vport-vxlan.c
net/openvswitch/vport.c
net/openvswitch/vport.h
The openvswitch conflicts were overlapping changes. One was
the egress tunnel info fix in 'net' and the other was the
vport ->send() op simplification in 'net-next'.
The xfrm6_output.c conflicts was also a simplification
overlapping a bug fix.
Signed-off-by: David S. Miller <davem@davemloft.net>
The current code for message reassembly is erroneously assuming that
the the first arriving fragment buffer always is linear, and then goes
ahead resetting the fragment list of that buffer in anticipation of
more arriving fragments.
However, if the buffer already happens to be non-linear, we will
inadvertently drop the already attached fragment list, and later
on trig a BUG() in __pskb_pull_tail().
We see this happen when running fragmented TIPC multicast across UDP,
something made possible since
commit d0f91938be ("tipc: add ip/udp media type")
We fix this by not resetting the fragment list when the buffer is non-
linear, and by initiatlizing our private fragment list tail pointer to
the tail of the existing fragment list.
Fixes: commit d0f91938be ("tipc: add ip/udp media type")
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After the previous commits, we are guaranteed that no packets
of type LINK_PROTOCOL or with illegal sequence numbers will be
attempted added to the link deferred queue. This makes it possible to
make some simplifications to the sorting algorithm in the function
tipc_skb_queue_sorted().
We also alter the function so that it will drop packets if one with
the same seqeunce number is already present in the queue. This is
necessary because we have identified weird packet sequences, involving
duplicate packets, where a legitimate in-sequence packet may advance to
the head of the queue without being detected and de-queued.
Finally, we make this function outline, since it will now be called only
in exceptional cases.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The msg pointer into header may change after skb linearization.
We must reinitialize it after calling skb_linearize to prevent
operating on a freed or invalid pointer.
Signed-off-by: Erik Hugne <erik.hugne@ericsson.com>
Reported-by: Tamás Végh <tamas.vegh@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a message is received in a socket, one of the call chains
tipc_sk_rcv()->tipc_sk_enqueue()->filter_rcv()(->tipc_sk_proto_rcv())
or
tipc_sk_backlog_rcv()->filter_rcv()(->tipc_sk_proto_rcv())
are followed. At each of these levels we may encounter situations
where the message may need to be rejected, or a new message
produced for transfer back to the sender. Despite recent
improvements, the current code for doing this is perceived
as awkward and hard to follow.
Leveraging the two previous commits in this series, we now
introduce a more uniform handling of such situations. We
let each of the functions in the chain itself produce/reverse
the message to be returned to the sender, but also perform the
actual forwarding. This simplifies the necessary logics within
each function.
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, we use the code sequence
if (msg_reverse())
tipc_link_xmit_skb()
at numerous locations in socket.c. The preparation of arguments
for these calls, as well as the sequence itself, makes the code
unecessarily complex.
In this commit, we introduce a new function, tipc_sk_respond(),
that performs this call combination. We also replace some, but not
yet all, of these explicit call sequences with calls to the new
function. Notably, we let the function tipc_sk_proto_rcv() use
the new function to directly send out PROBE_REPLY messages,
instead of deferring this to the calling tipc_sk_rcv() function,
as we do now.
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The shortest TIPC message header, for cluster local CONNECTED messages,
is 24 bytes long. With this format, the fields "dest_node" and
"orig_node" are optimized away, since they in reality are redundant
in this particular case.
However, the absence of these fields leads to code inconsistencies
that are difficult to handle in some cases, especially when we need
to reverse or reject messages at the socket layer.
In this commit, we concentrate the handling of the absent fields
to one place, by letting the function tipc_msg_reverse() reallocate
the buffer and expand the header to 32 bytes when necessary. This
means that the socket code now can assume that the two previously
absent fields are present in the header when a message needs to be
rejected. This opens up for some further simplifications of the
socket code.
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, the packet sequence number is updated and added to each
packet at the moment a packet is added to the link backlog queue.
This is wasteful, since it forces the code to traverse the send
packet list packet by packet when adding them to the backlog queue.
It would be better to just splice the whole packet list into the
backlog queue when that is the right action to do.
In this commit, we do this change. Also, since the sequence numbers
cannot now be assigned to the packets at the moment they are added
the backlog queue, we do instead calculate and add them at the moment
of transmission, when the backlog queue has to be traversed anyway.
We do this in the function tipc_link_push_packet().
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The link congestion algorithm used until now implies two problems.
- It is too generous towards lower-level messages in situations of high
load by giving "absolute" bandwidth guarantees to the different
priority levels. LOW traffic is guaranteed 10%, MEDIUM is guaranted
20%, HIGH is guaranteed 30%, and CRITICAL is guaranteed 40% of the
available bandwidth. But, in the absence of higher level traffic, the
ratio between two distinct levels becomes unreasonable. E.g. if there
is only LOW and MEDIUM traffic on a system, the former is guaranteed
1/3 of the bandwidth, and the latter 2/3. This again means that if
there is e.g. one LOW user and 10 MEDIUM users, the former will have
33.3% of the bandwidth, and the others will have to compete for the
remainder, i.e. each will end up with 6.7% of the capacity.
- Packets of type MSG_BUNDLER are created at SYSTEM importance level,
but only after the packets bundled into it have passed the congestion
test for their own respective levels. Since bundled packets don't
result in incrementing the level counter for their own importance,
only occasionally for the SYSTEM level counter, they do in practice
obtain SYSTEM level importance. Hence, the current implementation
provides a gap in the congestion algorithm that in the worst case
may lead to a link reset.
We now refine the congestion algorithm as follows:
- A message is accepted to the link backlog only if its own level
counter, and all superior level counters, permit it.
- The importance of a created bundle packet is set according to its
contents. A bundle packet created from messges at levels LOW to
CRITICAL is given importance level CRITICAL, while a bundle created
from a SYSTEM level message is given importance SYSTEM. In the latter
case only subsequent SYSTEM level messages are allowed to be bundled
into it.
This solves the first problem described above, by making the bandwidth
guarantee relative to the total number of users at all levels; only
the upper limit for each level remains absolute. In the example
described above, the single LOW user would use 1/11th of the bandwidth,
the same as each of the ten MEDIUM users, but he still has the same
guarantee against starvation as the latter ones.
The fix also solves the second problem. If the CRITICAL level is filled
up by bundle packets of that level, no lower level packets will be
accepted any more.
Suggested-by: Gergely Kiss <gergely.kiss@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a bearer is disabled manually, all its links have to be reset
and deleted. However, if there is a remaining, parallel link ready
to take over a deleted link's traffic, we currently delay the delete
of the removed link until the failover procedure is finished. This
is because the remaining link needs to access state from the reset
link, such as the last received packet number, and any partially
reassembled buffer, in order to perform a successful failover.
In this commit, we do instead move the state data over to the new
link, so that it can fulfill the procedure autonomously, without
accessing any data on the old link. This means that we can now
proceed and delete all pertaining links immediately when a bearer
is disabled. This saves us from some unnecessary complexity in such
situations.
We also choose to change the confusing definitions CHANGEOVER_PROTOCOL,
ORIGINAL_MSG and DUPLICATE_MSG to the more descriptive TUNNEL_PROTOCOL,
FAILOVER_MSG and SYNCH_MSG respectively.
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A message sent to a node after a successful name table lookup may still
find that the destination socket has disappeared, because distribution
of name table updates is non-atomic. If so, the message will be rejected
back to the sender with error code TIPC_ERR_NO_PORT. If the source
socket of the message has disappeared in the meantime, the message
should be dropped.
However, in the currrent code, the message will instead be subject to an
unwanted tertiary lookup, because the function tipc_msg_lookup_dest()
doesn't check if there is an error code present in the message before
performing the lookup. In the worst case, the message may now find the
old destination again, and be redirected once more, instead of being
dropped directly as it should be.
A second bug in this function is that the "prev_node" field in the message
is not updated after successful lookup, something that may have
unpredictable consequences.
The problems arising from those bugs occur very infrequently.
The third change in this function; the test on msg_reroute_msg_cnt() is
purely cosmetic, reflecting that the returned value never can be negative.
This commit corrects the two bugs described above.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
