Since OPTIONS_MPTCP_MPC has been defined, use it instead of open-coding.
Acked-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NGeliang Tang <geliangtang@xiaomi.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When recovering after a link failure, snd_nxt should not be set to a
lower value.  Else, update of snd_nxt is broken because:

  msk->snd_nxt += ret; (where ret is number of bytes sent)

assumes that snd_nxt always moves forward.
After reduction, its possible that snd_nxt update gets out of sync:
dfrag we just sent might have had a data sequence number even past
recovery_snd_nxt.

This change factors the common msk state update to a helper
and updates snd_nxt based on the current dfrag data sequence number.

The conditional is required for the recovery phase where we may
re-transmit old dfrags that are before current snd_nxt.

After this change, snd_nxt only moves forward and covers all in-sequence
data that was transmitted.

recovery_snd_nxt is retained to detect when recovery has completed.

Fixes: 1e1d9d6f ("mptcp: handle pending data on closed subflow")
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We want to revert the skb TX cache, but MPTCP is currently
using it unconditionally.

Rework the MPTCP tx code, so that tcp_tx_skb_cache is not
needed anymore: do the whole coalescing check, skb allocation
skb initialization/update inside mptcp_sendmsg_frag(), quite
alike the current TCP code.
Reviewed-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Due to signed/unsigned comparison, the expression:

	info->size_goal - skb->len > 0

evaluates to true when the size goal is smaller than the
skb size. That results in lack of tx cache refill, so that
the skb allocated by the core TCP code lacks the required
MPTCP skb extensions.

Due to the above, syzbot is able to trigger the following WARN_ON():

WARNING: CPU: 1 PID: 810 at net/mptcp/protocol.c:1366 mptcp_sendmsg_frag+0x1362/0x1bc0 net/mptcp/protocol.c:1366
Modules linked in:
CPU: 1 PID: 810 Comm: syz-executor.4 Not tainted 5.14.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:mptcp_sendmsg_frag+0x1362/0x1bc0 net/mptcp/protocol.c:1366
Code: ff 4c 8b 74 24 50 48 8b 5c 24 58 e9 0f fb ff ff e8 13 44 8b f8 4c 89 e7 45 31 ed e8 98 57 2e fe e9 81 f4 ff ff e8 fe 43 8b f8 <0f> 0b 41 bd ea ff ff ff e9 6f f4 ff ff 4c 89 e7 e8 b9 8e d2 f8 e9
RSP: 0018:ffffc9000531f6a0 EFLAGS: 00010216
RAX: 000000000000697f RBX: 0000000000000000 RCX: ffffc90012107000
RDX: 0000000000040000 RSI: ffffffff88eac9e2 RDI: 0000000000000003
RBP: ffff888078b15780 R08: 0000000000000000 R09: 0000000000000000
R10: ffffffff88eac017 R11: 0000000000000000 R12: ffff88801de0a280
R13: 0000000000006b58 R14: ffff888066278280 R15: ffff88803c2fe9c0
FS:  00007fd9f866e700(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007faebcb2f718 CR3: 00000000267cb000 CR4: 00000000001506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 __mptcp_push_pending+0x1fb/0x6b0 net/mptcp/protocol.c:1547
 mptcp_release_cb+0xfe/0x210 net/mptcp/protocol.c:3003
 release_sock+0xb4/0x1b0 net/core/sock.c:3206
 sk_stream_wait_memory+0x604/0xed0 net/core/stream.c:145
 mptcp_sendmsg+0xc39/0x1bc0 net/mptcp/protocol.c:1749
 inet6_sendmsg+0x99/0xe0 net/ipv6/af_inet6.c:643
 sock_sendmsg_nosec net/socket.c:704 [inline]
 sock_sendmsg+0xcf/0x120 net/socket.c:724
 sock_write_iter+0x2a0/0x3e0 net/socket.c:1057
 call_write_iter include/linux/fs.h:2163 [inline]
 new_sync_write+0x40b/0x640 fs/read_write.c:507
 vfs_write+0x7cf/0xae0 fs/read_write.c:594
 ksys_write+0x1ee/0x250 fs/read_write.c:647
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x4665f9
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fd9f866e188 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 000000000056c038 RCX: 00000000004665f9
RDX: 00000000000e7b78 RSI: 0000000020000000 RDI: 0000000000000003
RBP: 00000000004bfcc4 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 000000000056c038
R13: 0000000000a9fb1f R14: 00007fd9f866e300 R15: 0000000000022000

Fix the issue rewriting the relevant expression to avoid
sign-related problems - note: size_goal is always >= 0.

Additionally, ensure that the skb in the tx cache always carries
the relevant extension.

Reported-and-tested-by: syzbot+263a248eec3e875baa7b@syzkaller.appspotmail.com
Fixes: 1094c6fe ("mptcp: fix possible divide by zero")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Due to signed/unsigned comparison, the expression:

	info->size_goal - skb->len > 0

evaluates to true when the size goal is smaller than the
skb size. That results in lack of tx cache refill, so that
the skb allocated by the core TCP code lacks the required
MPTCP skb extensions.

Due to the above, syzbot is able to trigger the following WARN_ON():

WARNING: CPU: 1 PID: 810 at net/mptcp/protocol.c:1366 mptcp_sendmsg_frag+0x1362/0x1bc0 net/mptcp/protocol.c:1366
Modules linked in:
CPU: 1 PID: 810 Comm: syz-executor.4 Not tainted 5.14.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:mptcp_sendmsg_frag+0x1362/0x1bc0 net/mptcp/protocol.c:1366
Code: ff 4c 8b 74 24 50 48 8b 5c 24 58 e9 0f fb ff ff e8 13 44 8b f8 4c 89 e7 45 31 ed e8 98 57 2e fe e9 81 f4 ff ff e8 fe 43 8b f8 <0f> 0b 41 bd ea ff ff ff e9 6f f4 ff ff 4c 89 e7 e8 b9 8e d2 f8 e9
RSP: 0018:ffffc9000531f6a0 EFLAGS: 00010216
RAX: 000000000000697f RBX: 0000000000000000 RCX: ffffc90012107000
RDX: 0000000000040000 RSI: ffffffff88eac9e2 RDI: 0000000000000003
RBP: ffff888078b15780 R08: 0000000000000000 R09: 0000000000000000
R10: ffffffff88eac017 R11: 0000000000000000 R12: ffff88801de0a280
R13: 0000000000006b58 R14: ffff888066278280 R15: ffff88803c2fe9c0
FS:  00007fd9f866e700(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007faebcb2f718 CR3: 00000000267cb000 CR4: 00000000001506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 __mptcp_push_pending+0x1fb/0x6b0 net/mptcp/protocol.c:1547
 mptcp_release_cb+0xfe/0x210 net/mptcp/protocol.c:3003
 release_sock+0xb4/0x1b0 net/core/sock.c:3206
 sk_stream_wait_memory+0x604/0xed0 net/core/stream.c:145
 mptcp_sendmsg+0xc39/0x1bc0 net/mptcp/protocol.c:1749
 inet6_sendmsg+0x99/0xe0 net/ipv6/af_inet6.c:643
 sock_sendmsg_nosec net/socket.c:704 [inline]
 sock_sendmsg+0xcf/0x120 net/socket.c:724
 sock_write_iter+0x2a0/0x3e0 net/socket.c:1057
 call_write_iter include/linux/fs.h:2163 [inline]
 new_sync_write+0x40b/0x640 fs/read_write.c:507
 vfs_write+0x7cf/0xae0 fs/read_write.c:594
 ksys_write+0x1ee/0x250 fs/read_write.c:647
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x4665f9
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fd9f866e188 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 000000000056c038 RCX: 00000000004665f9
RDX: 00000000000e7b78 RSI: 0000000020000000 RDI: 0000000000000003
RBP: 00000000004bfcc4 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 000000000056c038
R13: 0000000000a9fb1f R14: 00007fd9f866e300 R15: 0000000000022000

Fix the issue rewriting the relevant expression to avoid
sign-related problems - note: size_goal is always >= 0.

Additionally, ensure that the skb in the tx cache always carries
the relevant extension.

Reported-and-tested-by: syzbot+263a248eec3e875baa7b@syzkaller.appspotmail.com
Fixes: 1094c6fe ("mptcp: fix possible divide by zero")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This retrieves the address pairs of all subflows currently
active for a given mptcp connection.

It re-uses the same meta-header as for MPTCP_TCPINFO.

A new structure is provided to hold the subflow
address data:

struct mptcp_subflow_addrs {
	union {
		__kernel_sa_family_t sa_family;
		struct sockaddr sa_local;
		struct sockaddr_in sin_local;
		struct sockaddr_in6 sin6_local;
		struct sockaddr_storage ss_local;
	};
	union {
		struct sockaddr sa_remote;
		struct sockaddr_in sin_remote;
		struct sockaddr_in6 sin6_remote;
		struct sockaddr_storage ss_remote;
	};
};

Usage of the new getsockopt is very similar to
MPTCP_TCPINFO one.

Userspace allocates a
'struct mptcp_subflow_data', followed by one or
more 'struct mptcp_subflow_addrs', then inits the
mptcp_subflow_data structure as follows:

struct mptcp_subflow_addrs *sf_addr;
struct mptcp_subflow_data *addr;
socklen_t olen = sizeof(*addr) + (8 * sizeof(*sf_addr));

addr = malloc(olen);
addr->size_subflow_data = sizeof(*addr);
addr->num_subflows = 0;
addr->size_kernel = 0;
addr->size_user = sizeof(struct mptcp_subflow_addrs);

sf_addr = (struct mptcp_subflow_addrs *)(addr + 1);

and then retrieves the endpoint addresses via:
ret = getsockopt(fd, SOL_MPTCP, MPTCP_SUBFLOW_ADDRS,
		 addr, &olen);

If the call succeeds, kernel will have added up to 8
endpoint addresses after the 'mptcp_subflow_data' header.

Userspace needs to re-check 'olen' value to detect how
many bytes have been filled in by the kernel.

Userspace can check addr->num_subflows to discover when
there were more subflows that available data space.
Co-developed-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Allow users to retrieve TCP_INFO data of all subflows.

Users need to pre-initialize a meta header that has to be
prepended to the data buffer that will be filled with the tcp info data.

The meta header looks like this:

struct mptcp_subflow_data {
 __u32 size_subflow_data;/* size of this structure in userspace */
 __u32 num_subflows;	/* must be 0, set by kernel */
 __u32 size_kernel;	/* must be 0, set by kernel */
 __u32 size_user;	/* size of one element in data[] */
} __attribute__((aligned(8)));

size_subflow_data has to be set to 'sizeof(struct mptcp_subflow_data)'.
This allows to extend mptcp_subflow_data structure later on without
breaking backwards compatibility.

If the structure is extended later on, kernel knows where the
userspace-provided meta header ends, even if userspace uses an older
(smaller) version of the structure.

num_subflows must be set to 0. If the getsockopt request succeeds (return
value is 0), it will be updated to contain the number of active subflows
for the given logical connection.

size_kernel must be set to 0. If the getsockopt request is successful,
it will contain the size of the 'struct tcp_info' as known by the kernel.
This is informational only.

size_user must be set to 'sizeof(struct tcp_info)'.

This allows the kernel to only fill in the space reserved/expected by
userspace.

Example:

struct my_tcp_info {
  struct mptcp_subflow_data d;
  struct tcp_info ti[2];
};
struct my_tcp_info ti;
socklen_t olen;

memset(&ti, 0, sizeof(ti));

ti.d.size_subflow_data = sizeof(struct mptcp_subflow_data);
ti.d.size_user = sizeof(struct tcp_info);
olen = sizeof(ti);

ret = getsockopt(fd, SOL_MPTCP, MPTCP_TCPINFO, &ti, &olen);
if (ret < 0)
	die_perror("getsockopt MPTCP_TCPINFO");

mptcp_subflow_data.num_subflows is populated with the number of
subflows that exist on the kernel side for the logical mptcp connection.

This allows userspace to re-try with a larger tcp_info array if the number
of subflows was larger than the available space in the ti[] array.

olen has to be set to the number of bytes that userspace has allocated to
receive the kernel data.  It will be updated to contain the real number
bytes that have been copied to by the kernel.

In the above example, if the number if subflows was 1, olen is equal to
'sizeof(struct mptcp_subflow_data) + sizeof(struct tcp_info).
For 2 or more subflows olen is equal to 'sizeof(struct my_tcp_info)'.

If there was more data that could not be copied due to lack of space
in the option buffer, userspace can detect this by checking
mptcp_subflow_data->num_subflows.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Its not compatible with multipath-tcp.org kernel one.

1. The out-of-tree implementation defines a different 'struct mptcp_info',
   with embedded __user addresses for additional data such as
   endpoint addresses.

2. Mat Martineau points out that embedded __user addresses doesn't work
with BPF_CGROUP_RUN_PROG_GETSOCKOPT() which assumes that copying in
optsize bytes from optval provides all data that got copied to userspace.

This provides mptcp_info data for the given mptcp socket.

Userspace sets optlen to the size of the structure it expects.
The kernel updates it to contain the number of bytes that it copied.

This allows to append more information to the structure later.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Will be re-used from getsockopt path.
Since diag can be a module, we can't export the helper from diag, it
needs to be moved to core.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Recent changes exposed a bug where specifically-timed requests to the
path manager netlink API could trigger a divide-by-zero in
__tcp_select_window(), as syzkaller does:

divide error: 0000 [#1] SMP KASAN NOPTI
CPU: 0 PID: 9667 Comm: syz-executor.0 Not tainted 5.14.0-rc6+ #3
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
RIP: 0010:__tcp_select_window+0x509/0xa60 net/ipv4/tcp_output.c:3016
Code: 44 89 ff e8 c9 29 e9 fd 45 39 e7 0f 8d 20 ff ff ff e8 db 28 e9 fd 44 89 e3 e9 13 ff ff ff e8 ce 28 e9 fd 44 89 e0 44 89 e3 99 <f7> 7c 24 04 29 d3 e9 fc fe ff ff e8 b7 28 e9 fd 44 89 f1 48 89 ea
RSP: 0018:ffff888031ccf020 EFLAGS: 00010216
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000040000
RDX: 0000000000000000 RSI: ffff88811532c080 RDI: 0000000000000002
RBP: 0000000000000000 R08: ffffffff835807c2 R09: 0000000000000000
R10: 0000000000000004 R11: ffffed1020b92441 R12: 0000000000000000
R13: 1ffff11006399e08 R14: 0000000000000000 R15: 0000000000000000
FS:  00007fa4c8344700(0000) GS:ffff88811ae00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000001b2f424000 CR3: 000000003e4e2003 CR4: 0000000000770ef0
PKRU: 55555554
Call Trace:
 tcp_select_window net/ipv4/tcp_output.c:264 [inline]
 __tcp_transmit_skb+0xc00/0x37a0 net/ipv4/tcp_output.c:1351
 __tcp_send_ack.part.0+0x3ec/0x760 net/ipv4/tcp_output.c:3972
 __tcp_send_ack net/ipv4/tcp_output.c:3978 [inline]
 tcp_send_ack+0x7d/0xa0 net/ipv4/tcp_output.c:3978
 mptcp_pm_nl_addr_send_ack+0x1ab/0x380 net/mptcp/pm_netlink.c:654
 mptcp_pm_remove_addr+0x161/0x200 net/mptcp/pm.c:58
 mptcp_nl_remove_id_zero_address+0x197/0x460 net/mptcp/pm_netlink.c:1328
 mptcp_nl_cmd_del_addr+0x98b/0xd40 net/mptcp/pm_netlink.c:1359
 genl_family_rcv_msg_doit.isra.0+0x225/0x340 net/netlink/genetlink.c:731
 genl_family_rcv_msg net/netlink/genetlink.c:775 [inline]
 genl_rcv_msg+0x341/0x5b0 net/netlink/genetlink.c:792
 netlink_rcv_skb+0x148/0x430 net/netlink/af_netlink.c:2504
 genl_rcv+0x24/0x40 net/netlink/genetlink.c:803
 netlink_unicast_kernel net/netlink/af_netlink.c:1314 [inline]
 netlink_unicast+0x537/0x750 net/netlink/af_netlink.c:1340
 netlink_sendmsg+0x846/0xd80 net/netlink/af_netlink.c:1929
 sock_sendmsg_nosec net/socket.c:704 [inline]
 sock_sendmsg+0x14e/0x190 net/socket.c:724
 ____sys_sendmsg+0x709/0x870 net/socket.c:2403
 ___sys_sendmsg+0xff/0x170 net/socket.c:2457
 __sys_sendmsg+0xe5/0x1b0 net/socket.c:2486
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x44/0xae

mptcp_pm_nl_addr_send_ack() was attempting to send a TCP ACK on the
first subflow in the MPTCP socket's connection list without validating
that the subflow was in a suitable connection state. To address this,
always validate subflow state when sending extra ACKs on subflows
for address advertisement or subflow priority change.

Fixes: 84dfe367 ("mptcp: send out dedicated ADD_ADDR packet")
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/229Co-developed-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Acked-by: NGeliang Tang <geliangtang@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Florian noted that if mptcp_alloc_tx_skb() allocation fails
in __mptcp_push_pending(), we can end-up invoking
mptcp_push_release()/tcp_push() with a zero mss, causing
a divide by 0 error.

This change addresses the issue refactoring the skb allocation
code checking if skb collapsing will happen for sure and doing
the skb allocation only after such check. Skb allocation will
now happen only after the call to tcp_send_mss() which
correctly initializes mss_now.

As side bonuses we now fill the skb tx cache only when needed,
and this also clean-up a bit the output path.

v1 -> v2:
 - use lockdep_assert_held_once() - Jakub
 - fix indentation - Jakub
Reported-by: NFlorian Westphal <fw@strlen.de>
Fixes: 724cfd2e ("mptcp: allocate TX skbs in msk context")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Fix the following coccicheck warning:
./net/mptcp/protocol.h:36:50-73: duplicated argument to & or |

The OPTION_MPTCP_MPJ_SYNACK here is duplicate.
Here should be OPTION_MPTCP_MPJ_ACK.

Fixes: 74c7dfbe ("mptcp: consolidate in_opt sub-options fields in a bitmask")
Signed-off-by: NWan Jiabing <wanjiabing@vivo.com>
Acked-by: NPaolo Abeni <pabeni@redhat.com>
Reviewed-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Florian noted the locking schema used by __mptcp_push_pending()
is hard to follow, let's add some more descriptive comments
and drop an unneeded and confusing check.
Suggested-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Most MPTCP packets carries a single MPTCP subption: the
DSS containing the mapping for the current packet.

Check explicitly for the above, so that is such scenario we
replace most conditional statements with a single likely() one.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This makes input options processing more consistent with
output ones and will simplify the next patch.

Also avoid clearing the suboption field after processing
it, since it's not needed.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change reorder the mptcp_options_received fields
to shrink the structure a bit and to ensure the most
frequently used fields are all in the first cacheline.

Sub-opt specific flags are moved out of the suboptions area,
and we must now explicitly set them when the relevant
suboption is parsed.

There is a notable exception: 'csum_reqd' is used by both DSS
and MPC suboptions, and keeping such field in the suboptions
flag area will simplfy the next patch.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Should be set only if the ingress packets present it, otherwise
we can confuse csum validation.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch added the mibs for MP_FAIL: MPTCP_MIB_MPFAILTX and
MPTCP_MIB_MPFAILRX.
Signed-off-by: NGeliang Tang <geliangtang@xiaomi.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a bad checksum is detected, set the send_mp_fail flag to send out
the MP_FAIL option.

Add a new function mptcp_has_another_subflow() to check whether there's
only a single subflow.

When multiple subflows are in use, close the affected subflow with a RST
that includes an MP_FAIL option and discard the data with the bad
checksum.

Set the sk_state of the subsocket to TCP_CLOSE, then the flag
MPTCP_WORK_CLOSE_SUBFLOW will be set in subflow_sched_work_if_closed,
and the subflow will be closed.

When a single subfow is in use, temporarily handled by sending MP_FAIL
with a RST too.
Signed-off-by: NGeliang Tang <geliangtang@xiaomi.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch added handling for receiving MP_FAIL suboption.

Add a new members mp_fail and fail_seq in struct mptcp_options_received.
When MP_FAIL suboption is received, set mp_fail to 1 and save the sequence
number to fail_seq.

Then invoke mptcp_pm_mp_fail_received to deal with the MP_FAIL suboption.
Signed-off-by: NGeliang Tang <geliangtang@xiaomi.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch added the MP_FAIL suboption sending support.

Add a new flag named send_mp_fail in struct mptcp_subflow_context. If
this flag is set, send out MP_FAIL suboption.

Add a new member fail_seq in struct mptcp_out_options to save the data
sequence number to put into the MP_FAIL suboption.

An MP_FAIL option could be included in a RST or on the subflow-level
ACK.
Suggested-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NGeliang Tang <geliangtang@xiaomi.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently we have several protocol constraints on MPTCP options
generation (e.g. MPC and MPJ subopt are mutually exclusive)
and some additional ones required by our implementation
(e.g. almost all ADD_ADDR variant are mutually exclusive with
everything else).

We can leverage the above to optimize the out option generation:
we check DSS/MPC/MPJ presence in a mutually exclusive way,
avoiding many unneeded conditionals in the common cases.

Additionally extend the existing constraints on ADD_ADDR opt on
all subvariants, so that it becomes fully mutually exclusive with
the above and we can skip another conditional statement for the
common case.

This change is also needed by the next patch.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

MPTCP_ADD_ADDR_IPV6 and MPTCP_ADD_ADDR_PORT are not necessary, we can get
these info from pm.local or pm.remote.

Drop mptcp_pm_should_add_signal_ipv6 and mptcp_pm_should_add_signal_port
too.
Co-developed-by: NGeliang Tang <geliangtang@gmail.com>
Signed-off-by: NGeliang Tang <geliangtang@gmail.com>
Signed-off-by: NYonglong Li <liyonglong@chinatelecom.cn>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

According to the MPTCP_ADD_ADDR_SIGNAL or MPTCP_ADD_ADDR_ECHO flag, build
the ADD_ADDR/ADD_ADDR_ECHO option.

In mptcp_pm_add_addr_signal(), use opts->addr to save the announced
ADD_ADDR or ADD_ADDR_ECHO address.
Co-developed-by: NGeliang Tang <geliangtang@gmail.com>
Signed-off-by: NGeliang Tang <geliangtang@gmail.com>
Co-developed-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NYonglong Li <liyonglong@chinatelecom.cn>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

ADD_ADDR shares pm.addr_signal with RM_ADDR, so after RM_ADDR/ADD_ADDR
has done, we should not clean ADD_ADDR/RM_ADDR's addr_signal.
Co-developed-by: NGeliang Tang <geliangtang@gmail.com>
Signed-off-by: NGeliang Tang <geliangtang@gmail.com>
Signed-off-by: NYonglong Li <liyonglong@chinatelecom.cn>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use MPTCP_ADD_ADDR_SIGNAL only for the action of sending ADD_ADDR, and
use MPTCP_ADD_ADDR_ECHO only for the action of sending ADD_ADDR echo.

Use msk->pm.local to save the announced ADD_ADDR address only, and reuse
msk->pm.remote to save the announced ADD_ADDR_ECHO address.

To prepare for the next patch.
Co-developed-by: NGeliang Tang <geliangtang@gmail.com>
Signed-off-by: NGeliang Tang <geliangtang@gmail.com>
Signed-off-by: NYonglong Li <liyonglong@chinatelecom.cn>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch moved the drop_other_suboptions check from
mptcp_established_options_add_addr() into mptcp_pm_add_addr_signal(), do
it under the PM lock to avoid the race between this check and
mptcp_pm_add_addr_signal().

For this, added a new parameter for mptcp_pm_add_addr_signal() to get
the drop_other_suboptions value. And drop the other suboptions after the
option length check if drop_other_suboptions is true.

Additionally, always drop the other suboption for TCP pure ack:
that makes both the code simpler and the MPTCP behaviour more
consistent.
Co-developed-by: NGeliang Tang <geliangtang@gmail.com>
Signed-off-by: NGeliang Tang <geliangtang@gmail.com>
Co-developed-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NYonglong Li <liyonglong@chinatelecom.cn>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If directly after an MP_CAPABLE 3WHS, the client receives an ADD_ADDR
with HMAC from the server, it is enough to switch to a "fully
established" mode because it has received more MPTCP options.

It was then OK to enable the "fully_established" flag on the MPTCP
socket. Still, best to check if the ADD_ADDR looks valid by looking if
it contains an HMAC (no 'echo' bit). If an ADD_ADDR echo is received
while we are not in "fully established" mode, it is strange and then
we should not switch to this mode now.

But that is not enough. On one hand, the path-manager has be notified
the state has changed. On the other hand, the "fully_established" flag
on the subflow socket should be turned on as well not to re-send the
MP_CAPABLE 3rd ACK content with the next ACK.

Fixes: 84dfe367 ("mptcp: send out dedicated ADD_ADDR packet")
Signed-off-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The endpoint cleanup path is prone to a memory leak, as reported
by syzkaller:

 BUG: memory leak
 unreferenced object 0xffff88810680ea00 (size 64):
   comm "syz-executor.6", pid 6191, jiffies 4295756280 (age 24.138s)
   hex dump (first 32 bytes):
     58 75 7d 3c 80 88 ff ff 22 01 00 00 00 00 ad de  Xu}<....".......
     01 00 02 00 00 00 00 00 ac 1e 00 07 00 00 00 00  ................
   backtrace:
     [<0000000072a9f72a>] kmalloc include/linux/slab.h:591 [inline]
     [<0000000072a9f72a>] mptcp_nl_cmd_add_addr+0x287/0x9f0 net/mptcp/pm_netlink.c:1170
     [<00000000f6e931bf>] genl_family_rcv_msg_doit.isra.0+0x225/0x340 net/netlink/genetlink.c:731
     [<00000000f1504a2c>] genl_family_rcv_msg net/netlink/genetlink.c:775 [inline]
     [<00000000f1504a2c>] genl_rcv_msg+0x341/0x5b0 net/netlink/genetlink.c:792
     [<0000000097e76f6a>] netlink_rcv_skb+0x148/0x430 net/netlink/af_netlink.c:2504
     [<00000000ceefa2b8>] genl_rcv+0x24/0x40 net/netlink/genetlink.c:803
     [<000000008ff91aec>] netlink_unicast_kernel net/netlink/af_netlink.c:1314 [inline]
     [<000000008ff91aec>] netlink_unicast+0x537/0x750 net/netlink/af_netlink.c:1340
     [<0000000041682c35>] netlink_sendmsg+0x846/0xd80 net/netlink/af_netlink.c:1929
     [<00000000df3aa8e7>] sock_sendmsg_nosec net/socket.c:704 [inline]
     [<00000000df3aa8e7>] sock_sendmsg+0x14e/0x190 net/socket.c:724
     [<000000002154c54c>] ____sys_sendmsg+0x709/0x870 net/socket.c:2403
     [<000000001aab01d7>] ___sys_sendmsg+0xff/0x170 net/socket.c:2457
     [<00000000fa3b1446>] __sys_sendmsg+0xe5/0x1b0 net/socket.c:2486
     [<00000000db2ee9c7>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
     [<00000000db2ee9c7>] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80
     [<000000005873517d>] entry_SYSCALL_64_after_hwframe+0x44/0xae

We should not require an allocation to cleanup stuff.

Rework the code a bit so that the additional RCU work is no more needed.

Fixes: 1729cf18 ("mptcp: create the listening socket for new port")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In mptcp_pm_nl_add_addr_received(), fill a temporary allocate array of
all local address corresponding to the fullmesh endpoint. If such array
is empty, keep the current behavior.

Elsewhere loop on such array and create a subflow for each local address
towards the given remote address
Suggested-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NGeliang Tang <geliangtang@xiaomi.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch added and managed a new per endpoint flag, named
MPTCP_PM_ADDR_FLAG_FULLMESH.

In mptcp_pm_create_subflow_or_signal_addr(), if such flag is set, instead
of:
        remote_address((struct sock_common *)sk, &remote);
fill a temporary allocated array of all known remote address. After
releaseing the pm lock loop on such array and create a subflow for each
remote address from the given local.

Note that the we could still use an array even for non 'fullmesh'
endpoint: with a single entry corresponding to the primary MPC subflow
remote address.
Suggested-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NGeliang Tang <geliangtang@xiaomi.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch added a new helper mptcp_pm_get_flags_and_ifindex_by_id(),
and used it in __mptcp_subflow_connect() to get the flags and ifindex
values.

Then the two arguments flags and ifindex of __mptcp_subflow_connect()
can be dropped.
Signed-off-by: NGeliang Tang <geliangtang@xiaomi.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

the parsed incoming backup flag is not propagated
to the subflow itself, the client may end-up using it
to send data.

Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/191Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This allows monitoring exceptional events like
active backup scenarios.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The msk can use backup subflows to transmit in-sequence data
only if there are no other active subflow. On active backup
scenario, the MPTCP connection can do forward progress only
due to MPTCP retransmissions - rtx can pick backup subflows.

This patch introduces a new flag flow MPTCP subflows: if the
underlying TCP connection made no progresses for long time,
and there are other less problematic subflows available, the
given subflow become stale.

Stale subflows are not considered active: if all non backup
subflows become stale, the MPTCP scheduler can pick backup
subflows for plain transmissions.

Stale subflows can return in active state, as soon as any reply
from the peer is observed.

Active backup scenarios can now leverage the available b/w
with no restrinction.

Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/207Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Reorder the data in mptcp_pernet to avoid wasting space
with no reasons and constify the access helpers.

No functional changes intended.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The PM can close active subflow, e.g. due to ingress RM_ADDR
option. Such subflow could carry data still unacked at the
MPTCP-level, both in the write and the rtx_queue, which has
never reached the other peer.

Currently the mptcp-level retransmission will deliver such data,
but at a very low rate (at most 1 DSM for each MPTCP rtx interval).

We can speed-up the recovery a lot, moving all the unacked in the
tcp write_queue, so that it will be pushed again via other
subflows, at the speed allowed by them.

Also make available the new helper for later patches.

Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/207Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current mptcp re-inject strategy is very aggressive,
we have mptcp-level retransmissions even on single subflow
connection, if the link in-use is lossy.

Let's be a little more conservative: we do retransmit
only if at least a subflow has write and rtx queue empty.

Additionally use the backup subflows only if the active
subflows are stale - no progresses in at least an rtx period
and ignore stale subflows for rtx timeout update

Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/207Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As reported by Maxim, we have a lot of MPTCP-level
retransmissions when multilple links with different latencies
are in use.

This patch refactor the mptcp-level timeout accounting so that
the maximum of all the active subflow timeout is used. To avoid
traversing the subflow list multiple times, the update is
performed inside the packet scheduler.

Additionally clean-up a bit timeout handling.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

kfree_rcu() had been removed from pm_netlink.c, so this rcu field in
struct mptcp_pm_addr_entry became useless. Let's drop it.
Signed-off-by: NGeliang Tang <geliangtang@gmail.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Link: https://lore.kernel.org/r/20210802231914.54709-1-mathew.j.martineau@linux.intel.comSigned-off-by: NJakub Kicinski <kuba@kernel.org>