| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
An SMC link group is bound to a specific vlan_id. Its link uses
the RoCE-GIDs established for the specific vlan_id. This patch makes
sure the appropriate vlan_id is determined for stacked scenarios like
for instance a master bonding device with vlan devices enslaved.
Signed-off-by: Ursula Braun <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
SIOCINQ returns the amount of unread data in the RMB.
SIOCOUTQ returns the amount of unsent or unacked sent data in the send
buffer.
SIOCOUTQNSD returns the amount of data prepared for sending, but
not yet sent.
Signed-off-by: Ursula Braun <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Update smc_diag.c to support ipv6 addresses on the diagnosis interface.
Signed-off-by: Karsten Graul <[email protected]>
Signed-off-by: Ursula Braun <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Add periodic LLC testlink support to ensure the link is still active.
The interval time is initialized using the value of
sysctl_tcp_keepalive_time.
Signed-off-by: Karsten Graul <[email protected]>
Signed-off-by: Ursula Braun <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
The smc_poll code tries to finish connect() if the socket is in
state SMC_INIT and polling of the internal CLC-socket returns with
EPOLLOUT. This makes sense for a select/poll call following a connect
call, but not without preceding connect().
With this patch smc_poll starts connect logic only, if the CLC-socket
is no longer in its initial state TCP_CLOSE.
In addition, a poll error on the internal CLC-socket is always
propagated to the SMC socket.
With this patch the code path mentioned by syzbot
https://syzkaller.appspot.com/bug?extid=03faa2dc16b8b64be396
is no longer possible.
Signed-off-by: Ursula Braun <[email protected]>
Reported-by: [email protected]
Signed-off-by: David S. Miller <[email protected]>
|
|
When auth is enabled for cookie-ack chunk, in sctp_inq_pop, sctp
processes auth chunk first, then continues to the next chunk in
this packet if chunk_end + chunk_hdr size < skb_tail_pointer().
Otherwise, it will go to the next packet or discard this chunk.
However, it missed the fact that cookie-ack chunk's size is equal
to chunk_hdr size, which couldn't match that check, and thus this
chunk would not get processed.
This patch fixes it by changing the check to chunk_end + chunk_hdr
size <= skb_tail_pointer().
Fixes: 26b87c788100 ("net: sctp: fix remote memory pressure from excessive queueing")
Signed-off-by: Xin Long <[email protected]>
Acked-by: Neil Horman <[email protected]>
Acked-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
When processing a duplicate cookie-echo chunk, for case 'D', sctp will
not process the param from this chunk. It means old asoc has nothing
to be updated, and the new temp asoc doesn't have the complete info.
So there's no reason to use the new asoc when creating the cookie-ack
chunk. Otherwise, like when auth is enabled for cookie-ack, the chunk
can not be set with auth, and it will definitely be dropped by peer.
This issue is there since very beginning, and we fix it by using the
old asoc instead.
Signed-off-by: Xin Long <[email protected]>
Acked-by: Neil Horman <[email protected]>
Acked-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
When processing a duplicate cookie-echo chunk, for case 'A' and 'B',
after sctp_process_init for the new asoc, if auth is enabled for the
cookie-ack chunk, the active key should also be initialized.
Otherwise, the cookie-ack chunk made later can not be set with auth
shkey properly, and a crash can even be caused by this, as after
Commit 1b1e0bc99474 ("sctp: add refcnt support for sh_key"), sctp
needs to hold the shkey when making control chunks.
Fixes: 1b1e0bc99474 ("sctp: add refcnt support for sh_key")
Reported-by: Jianwen Ji <[email protected]>
Signed-off-by: Xin Long <[email protected]>
Acked-by: Neil Horman <[email protected]>
Acked-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Previously the bbr->idle_restart tracking was zeroing out the
bbr->idle_restart bit upon ACKs that did not SACK or ACK anything,
e.g. receiving incoming data or receiver window updates. In such
situations BBR would forget that this was a restart-from-idle
situation, and if the min_rtt had expired it would unnecessarily enter
PROBE_RTT (even though we were actually restarting from idle but had
merely forgotten that fact).
The fix is simple: we need to remember we are restarting from idle
until we receive a S/ACK for some data (a S/ACK for the first flight
of data we send as we are restarting).
This commit is a stable candidate for kernels back as far as 4.9.
Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control")
Signed-off-by: Neal Cardwell <[email protected]>
Signed-off-by: Yuchung Cheng <[email protected]>
Signed-off-by: Soheil Hassas Yeganeh <[email protected]>
Signed-off-by: Priyaranjan Jha <[email protected]>
Signed-off-by: Yousuk Seung <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Using the udp_v4_check() function to calculate the pseudo header
for the newly segmented UDP packets results in assigning the complement
of the value to the UDP header checksum field.
Always undo the complement the partial checksum value in order to
match the case where GSO is not used on the UDP transmit path.
Fixes: ee80d1ebe5ba ("udp: add udp gso")
Signed-off-by: Sean Tranchetti <[email protected]>
Signed-off-by: Subash Abhinov Kasiviswanathan <[email protected]>
Acked-by: Willem de Bruijn <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
It is reported that in some cases, write_space may be called in
do_tcp_sendpages, such that we recursively invoke do_tcp_sendpages again:
[ 660.468802] ? do_tcp_sendpages+0x8d/0x580
[ 660.468826] ? tls_push_sg+0x74/0x130 [tls]
[ 660.468852] ? tls_push_record+0x24a/0x390 [tls]
[ 660.468880] ? tls_write_space+0x6a/0x80 [tls]
...
tls_push_sg already does a loop over all sending sg's, so ignore
any tls_write_space notifications until we are done sending.
We then have to call the previous write_space to wake up
poll() waiters after we are done with the send loop.
Reported-by: Andre Tomt <[email protected]>
Signed-off-by: Dave Watson <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Applications with many concurrent connections, high variance
in receive queue length and tight memory bounds cannot
allocate worst-case buffer size to drain sockets. Knowing
the size of receive queue length, applications can optimize
how they allocate buffers to read from the socket.
The number of bytes pending on the socket is directly
available through ioctl(FIONREAD/SIOCINQ) and can be
approximated using getsockopt(MEMINFO) (rmem_alloc includes
skb overheads in addition to application data). But, both of
these options add an extra syscall per recvmsg. Moreover,
ioctl(FIONREAD/SIOCINQ) takes the socket lock.
Add the TCP_INQ socket option to TCP. When this socket
option is set, recvmsg() relays the number of bytes available
on the socket for reading to the application via the
TCP_CM_INQ control message.
Calculate the number of bytes after releasing the socket lock
to include the processed backlog, if any. To avoid an extra
branch in the hot path of recvmsg() for this new control
message, move all cmsg processing inside an existing branch for
processing receive timestamps. Since the socket lock is not held
when calculating the size of receive queue, TCP_INQ is a hint.
For example, it can overestimate the queue size by one byte,
if FIN is received.
With this method, applications can start reading from the socket
using a small buffer, and then use larger buffers based on the
remaining data when needed.
V3 change-log:
As suggested by David Miller, added loads with barrier
to check whether we have multiple threads calling recvmsg
in parallel. When that happens we lock the socket to
calculate inq.
V4 change-log:
Removed inline from a static function.
Signed-off-by: Soheil Hassas Yeganeh <[email protected]>
Signed-off-by: Yuchung Cheng <[email protected]>
Signed-off-by: Willem de Bruijn <[email protected]>
Reviewed-by: Eric Dumazet <[email protected]>
Reviewed-by: Neal Cardwell <[email protected]>
Suggested-by: David Miller <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
We do not require this inline function to be used in multiple different
locations, just inline it where it gets used in register_netdevice().
Suggested-by: David Miller <[email protected]>
Suggested-by: Stephen Hemminger <[email protected]>
Signed-off-by: Florian Fainelli <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
It is valid to have static routes where the nexthop
is an interface not an address such as tunnels.
For IPv4 it was possible to use ECMP on these routes
but not for IPv6.
Signed-off-by: Thomas Winter <[email protected]>
Cc: David Ahern <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Alexey Kuznetsov <[email protected]>
Cc: Hideaki YOSHIFUJI <[email protected]>
Acked-by: David Ahern <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Syzbot managed to send a udp gso packet without checksum offload into
the gso stack by disabling tx checksum (UDP_NO_CHECK6_TX). This
triggered the skb_warn_bad_offload.
RIP: 0010:skb_warn_bad_offload+0x2bc/0x600 net/core/dev.c:2658
skb_gso_segment include/linux/netdevice.h:4038 [inline]
validate_xmit_skb+0x54d/0xd90 net/core/dev.c:3120
__dev_queue_xmit+0xbf8/0x34c0 net/core/dev.c:3577
dev_queue_xmit+0x17/0x20 net/core/dev.c:3618
UDP_NO_CHECK6_TX sets skb->ip_summed to CHECKSUM_NONE just after the
udp gso integrity checks in udp_(v6_)send_skb. Extend those checks to
catch and fail in this case.
After the integrity checks jump directly to the CHECKSUM_PARTIAL case
to avoid reading the no_check_tx flags again (a TOCTTOU race).
Fixes: bec1f6f69736 ("udp: generate gso with UDP_SEGMENT")
Signed-off-by: Willem de Bruijn <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
In ethtool_get_rxnfc(), the object "info" is firstly copied from
user-space. If the FLOW_RSS flag is set in the member field flow_type of
"info" (and cmd is ETHTOOL_GRXFH), info needs to be copied again from
user-space because FLOW_RSS is newer and has new definition, as mentioned
in the comment. However, given that the user data resides in user-space, a
malicious user can race to change the data after the first copy. By doing
so, the user can inject inconsistent data. For example, in the second
copy, the FLOW_RSS flag could be cleared in the field flow_type of "info".
In the following execution, "info" will be used in the function
ops->get_rxnfc(). Such inconsistent data can potentially lead to unexpected
information leakage since ops->get_rxnfc() will prepare various types of
data according to flow_type, and the prepared data will be eventually
copied to user-space. This inconsistent data may also cause undefined
behaviors based on how ops->get_rxnfc() is implemented.
This patch simply re-verifies the flow_type field of "info" after the
second copy. If the value is not as expected, an error code will be
returned.
Signed-off-by: Wenwen Wang <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Currently flower doesn't support inserting filters with different masks
on a single priority, even if the actual flows (key + mask) inserted
aren't overlapping, as with the use case of offloading openvswitch
datapath flows. Instead one must go up one level, and assign different
priorities for each mask, which will create a different flower
instances.
This patch opens flower to support more than one mask per priority,
and a single flower instance. It does so by adding another hash table
on top of the existing one which will store the different masks,
and the filters that share it.
The user is left with the responsibility of ensuring non overlapping
flows, otherwise precedence is not guaranteed.
Signed-off-by: Paul Blakey <[email protected]>
Signed-off-by: Jiri Pirko <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
syzbot is able to produce a nasty WARN_ON() in tcp_verify_left_out()
with following C-repro :
socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 3
setsockopt(3, SOL_TCP, TCP_REPAIR, [1], 4) = 0
setsockopt(3, SOL_TCP, TCP_REPAIR_QUEUE, [-1], 4) = 0
bind(3, {sa_family=AF_INET, sin_port=htons(20002), sin_addr=inet_addr("0.0.0.0")}, 16) = 0
sendto(3, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"...,
1242, MSG_FASTOPEN, {sa_family=AF_INET, sin_port=htons(20002), sin_addr=inet_addr("127.0.0.1")}, 16) = 1242
setsockopt(3, SOL_TCP, TCP_REPAIR_WINDOW, "\4\0\0@+\205\0\0\377\377\0\0\377\377\377\177\0\0\0\0", 20) = 0
writev(3, [{"\270", 1}], 1) = 1
setsockopt(3, SOL_TCP, TCP_REPAIR_OPTIONS, "\10\0\0\0\0\0\0\0\0\0\0\0|\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 386) = 0
writev(3, [{"\210v\r[\226\320t\231qwQ\204\264l\254\t\1\20\245\214p\350H\223\254;\\\37\345\307p$"..., 3144}], 1) = 3144
The 3rd system call looks odd :
setsockopt(3, SOL_TCP, TCP_REPAIR_QUEUE, [-1], 4) = 0
This patch makes sure bound checking is using an unsigned compare.
Fixes: ee9952831cfd ("tcp: Initial repair mode")
Signed-off-by: Eric Dumazet <[email protected]>
Reported-by: syzbot <[email protected]>
Cc: Pavel Emelyanov <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
syzbot/KMSAN reported an uninit-value in ip6_multipath_l3_keys(),
root caused to a bad assumption of ICMP header being already
pulled in skb->head
ip_multipath_l3_keys() does the correct thing, so it is an IPv6 only bug.
BUG: KMSAN: uninit-value in ip6_multipath_l3_keys net/ipv6/route.c:1830 [inline]
BUG: KMSAN: uninit-value in rt6_multipath_hash+0x5c4/0x640 net/ipv6/route.c:1858
CPU: 0 PID: 4507 Comm: syz-executor661 Not tainted 4.16.0+ #87
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
__dump_stack lib/dump_stack.c:17 [inline]
dump_stack+0x185/0x1d0 lib/dump_stack.c:53
kmsan_report+0x142/0x240 mm/kmsan/kmsan.c:1067
__msan_warning_32+0x6c/0xb0 mm/kmsan/kmsan_instr.c:683
ip6_multipath_l3_keys net/ipv6/route.c:1830 [inline]
rt6_multipath_hash+0x5c4/0x640 net/ipv6/route.c:1858
ip6_route_input+0x65a/0x920 net/ipv6/route.c:1884
ip6_rcv_finish+0x413/0x6e0 net/ipv6/ip6_input.c:69
NF_HOOK include/linux/netfilter.h:288 [inline]
ipv6_rcv+0x1e16/0x2340 net/ipv6/ip6_input.c:208
__netif_receive_skb_core+0x47df/0x4a90 net/core/dev.c:4562
__netif_receive_skb net/core/dev.c:4627 [inline]
netif_receive_skb_internal+0x49d/0x630 net/core/dev.c:4701
netif_receive_skb+0x230/0x240 net/core/dev.c:4725
tun_rx_batched drivers/net/tun.c:1555 [inline]
tun_get_user+0x740f/0x7c60 drivers/net/tun.c:1962
tun_chr_write_iter+0x1d4/0x330 drivers/net/tun.c:1990
call_write_iter include/linux/fs.h:1782 [inline]
new_sync_write fs/read_write.c:469 [inline]
__vfs_write+0x7fb/0x9f0 fs/read_write.c:482
vfs_write+0x463/0x8d0 fs/read_write.c:544
SYSC_write+0x172/0x360 fs/read_write.c:589
SyS_write+0x55/0x80 fs/read_write.c:581
do_syscall_64+0x309/0x430 arch/x86/entry/common.c:287
entry_SYSCALL_64_after_hwframe+0x3d/0xa2
Fixes: 23aebdacb05d ("ipv6: Compute multipath hash for ICMP errors from offending packet")
Signed-off-by: Eric Dumazet <[email protected]>
Reported-by: syzbot <[email protected]>
Cc: Jakub Sitnicki <[email protected]>
Acked-by: Jakub Sitnicki <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
The idea is quite similar to the old functions, but note that the _fixed
function wasn't "fixed" as in that it would generate a packet with a fixed
size, but rather limited/bounded to PMTU.
Also, now with sctp_mtu_payload(), we have a more accurate limit.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
And do so if the skb doesn't have enough space for the payload.
This is a preparation for the next patch.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This patch adds a generic infrastructure to offload TLS crypto to a
network device. It enables the kernel TLS socket to skip encryption
and authentication operations on the transmit side of the data path.
Leaving those computationally expensive operations to the NIC.
The NIC offload infrastructure builds TLS records and pushes them to
the TCP layer just like the SW KTLS implementation and using the same
API.
TCP segmentation is mostly unaffected. Currently the only exception is
that we prevent mixed SKBs where only part of the payload requires
offload. In the future we are likely to add a similar restriction
following a change cipher spec record.
The notable differences between SW KTLS and NIC offloaded TLS
implementations are as follows:
1. The offloaded implementation builds "plaintext TLS record", those
records contain plaintext instead of ciphertext and place holder bytes
instead of authentication tags.
2. The offloaded implementation maintains a mapping from TCP sequence
number to TLS records. Thus given a TCP SKB sent from a NIC offloaded
TLS socket, we can use the tls NIC offload infrastructure to obtain
enough context to encrypt the payload of the SKB.
A TLS record is released when the last byte of the record is ack'ed,
this is done through the new icsk_clean_acked callback.
The infrastructure should be extendable to support various NIC offload
implementations. However it is currently written with the
implementation below in mind:
The NIC assumes that packets from each offloaded stream are sent as
plaintext and in-order. It keeps track of the TLS records in the TCP
stream. When a packet marked for offload is transmitted, the NIC
encrypts the payload in-place and puts authentication tags in the
relevant place holders.
The responsibility for handling out-of-order packets (i.e. TCP
retransmission, qdisc drops) falls on the netdev driver.
The netdev driver keeps track of the expected TCP SN from the NIC's
perspective. If the next packet to transmit matches the expected TCP
SN, the driver advances the expected TCP SN, and transmits the packet
with TLS offload indication.
If the next packet to transmit does not match the expected TCP SN. The
driver calls the TLS layer to obtain the TLS record that includes the
TCP of the packet for transmission. Using this TLS record, the driver
posts a work entry on the transmit queue to reconstruct the NIC TLS
state required for the offload of the out-of-order packet. It updates
the expected TCP SN accordingly and transmits the now in-order packet.
The same queue is used for packet transmission and TLS context
reconstruction to avoid the need for flushing the transmit queue before
issuing the context reconstruction request.
Signed-off-by: Ilya Lesokhin <[email protected]>
Signed-off-by: Boris Pismenny <[email protected]>
Signed-off-by: Aviad Yehezkel <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
In TLS inline crypto, we can have one direction in software
and another in hardware. Thus, we split the TLS configuration to separate
structures for receive and transmit.
Signed-off-by: Boris Pismenny <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This patch adds a netdev feature to configure TLS TX offloads.
Signed-off-by: Ilya Lesokhin <[email protected]>
Signed-off-by: Boris Pismenny <[email protected]>
Signed-off-by: Aviad Yehezkel <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
With socket dependent offloads we rely on the netdev to transform
the transmitted packets before sending them to the wire.
When a packet from an offloaded socket is rerouted to a different
device we need to detect it and do the transformation in software.
Signed-off-by: Ilya Lesokhin <[email protected]>
Signed-off-by: Boris Pismenny <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
copy_skb_header is renamed to skb_copy_header and
exported. Exposing this function give more flexibility
in copying SKBs.
skb_copy and skb_copy_expand do not give enough control
over which parts are copied.
Signed-off-by: Ilya Lesokhin <[email protected]>
Signed-off-by: Boris Pismenny <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Called when a TCP segment is acknowledged.
Could be used by application protocols who hold additional
metadata associated with the stream data.
This is required by TLS device offload to release
metadata associated with acknowledged TLS records.
Signed-off-by: Ilya Lesokhin <[email protected]>
Signed-off-by: Boris Pismenny <[email protected]>
Signed-off-by: Aviad Yehezkel <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Add a couple new functions to allow querying FDB and vlan settings of a
bridge.
Signed-off-by: Petr Machata <[email protected]>
Signed-off-by: Ido Schimmel <[email protected]>
Signed-off-by: Nikolay Aleksandrov <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
The seg6_make_flowlabel() is used by seg6_do_srh_encap() to compute the
flowlabel from a given skb. It relies on skb_get_hash() which eventually
calls __skb_flow_dissect() to extract the flow_keys struct values from
the skb.
In case of IPv4 traffic, calling seg6_make_flowlabel() after skb_push(),
skb_reset_network_header(), and skb_mac_header_rebuild() will results in
flow_keys struct of all key values set to zero.
This patch calls seg6_make_flowlabel() before resetting the headers of skb
to get the right key values.
Extracted Key values are based on the type inner packet as follows:
1) IPv6 traffic: src_IP, dst_IP, L4 proto, and flowlabel of inner packet.
2) IPv4 traffic: src_IP, dst_IP, L4 proto, src_port, and dst_port
3) L2 traffic: depends on what kind of traffic carried into the L2
frame. IPv6 and IPv4 traffic works as discussed 1) and 2)
Here a hex_dump of struct flow_keys for IPv4 and IPv6 traffic
10.100.1.100: 47302 > 30.0.0.2: 5001
00000000: 14 00 02 00 00 00 00 00 08 00 11 00 00 00 00 00
00000010: 00 00 00 00 00 00 00 00 13 89 b8 c6 1e 00 00 02
00000020: 0a 64 01 64
fc00:a1:a > b2::2
00000000: 28 00 03 00 00 00 00 00 86 dd 11 00 99 f9 02 00
00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 b2 00 00
00000020: 00 00 00 00 00 00 00 00 00 00 00 02 fc 00 00 a1
00000030: 00 00 00 00 00 00 00 00 00 00 00 0a
Signed-off-by: Ahmed Abdelsalam <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Currently the truncated bit is set only when the mirrored packet
is larger than mtu. For certain cases, the packet might already
been truncated before sending to the erspan tunnel. In this case,
the patch detect whether the IP header's total length is larger
than the actual skb->len. If true, this indicated that the
mirrored packet is truncated and set the erspan truncate bit.
I tested the patch using bpf_skb_change_tail helper function to
shrink the packet size and send to erspan tunnel.
Reported-by: Xiaoyan Jin <[email protected]>
Signed-off-by: William Tu <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
We have about 53 netdev_features_t bits defined and counting, add a
build time check to catch when an u64 type will not be enough and we
will have to convert that to a bitmap. This is done in
register_netdevice() for convenience.
Signed-off-by: Florian Fainelli <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
I am dropping the export of __skb_tx_hash as after my patches nobody is
using it outside of the net/core/dev.c file. In addition I am renaming and
repurposing it to just be a static declaration of skb_tx_hash since that
was the only user for it at this point. By doing this the compiler can
inline it into __netdev_pick_tx as that will improve performance.
Signed-off-by: Alexander Duyck <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
When adding tcp mmap() implementation, I forgot that socket lock
had to be taken before current->mm->mmap_sem. syzbot eventually caught
the bug.
Since we can not lock the socket in tcp mmap() handler we have to
split the operation in two phases.
1) mmap() on a tcp socket simply reserves VMA space, and nothing else.
This operation does not involve any TCP locking.
2) getsockopt(fd, IPPROTO_TCP, TCP_ZEROCOPY_RECEIVE, ...) implements
the transfert of pages from skbs to one VMA.
This operation only uses down_read(¤t->mm->mmap_sem) after
holding TCP lock, thus solving the lockdep issue.
This new implementation was suggested by Andy Lutomirski with great details.
Benefits are :
- Better scalability, in case multiple threads reuse VMAS
(without mmap()/munmap() calls) since mmap_sem wont be write locked.
- Better error recovery.
The previous mmap() model had to provide the expected size of the
mapping. If for some reason one part could not be mapped (partial MSS),
the whole operation had to be aborted.
With the tcp_zerocopy_receive struct, kernel can report how
many bytes were successfuly mapped, and how many bytes should
be read to skip the problematic sequence.
- No more memory allocation to hold an array of page pointers.
16 MB mappings needed 32 KB for this array, potentially using vmalloc() :/
- skbs are freed while mmap_sem has been released
Following patch makes the change in tcp_mmap tool to demonstrate
one possible use of mmap() and setsockopt(... TCP_ZEROCOPY_RECEIVE ...)
Note that memcg might require additional changes.
Fixes: 93ab6cc69162 ("tcp: implement mmap() for zero copy receive")
Signed-off-by: Eric Dumazet <[email protected]>
Reported-by: syzbot <[email protected]>
Suggested-by: Andy Lutomirski <[email protected]>
Cc: [email protected]
Acked-by: Soheil Hassas Yeganeh <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
When we set a bond slave's master to bridge via ioctl, we only check
the IFF_BRIDGE_PORT flag. Although we will find the slave's real master
at netdev_master_upper_dev_link() later, it already does some settings
and allocates some resources. It would be better to return as early
as possible.
v1 -> v2:
use netdev_master_upper_dev_get() instead of netdev_has_any_upper_dev()
to check if we have a master, because not all upper devs are masters,
e.g. vlan device.
Reported-by: [email protected]
Signed-off-by: Hangbin Liu <[email protected]>
Acked-by: Nikolay Aleksandrov <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
UDP GSO needs to export __udp_gso_segment to call it from ipv6.
I accidentally exported static ipv4 function __udp4_gso_segment.
Remove that EXPORT_SYMBOL_GPL.
Fixes: ee80d1ebe5ba ("udp: add udp gso")
Signed-off-by: Willem de Bruijn <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
For the x32 ABI, struct timeval has two 64-bit fields. However
the kernel currently interprets the user-space values used for
the SO_RCVTIMEO and SO_SNDTIMEO socket options as having a pair
of 32-bit fields.
When the seconds portion of the requested timeout is less than 2**32,
the seconds portion of the effective timeout is correct but the
microseconds portion is zero. When the seconds portion of the
requested timeout is zero and the microseconds portion is non-zero,
the kernel interprets the timeout as zero (never timeout).
Fix by using 64-bit time for SO_RCVTIMEO/SO_SNDTIMEO as required
for the ABI.
The code included below demonstrates the problem.
Results before patch:
$ gcc -m64 -Wall -O2 -o socktmo socktmo.c && ./socktmo
recv time: 2.008181 seconds
send time: 2.015985 seconds
$ gcc -m32 -Wall -O2 -o socktmo socktmo.c && ./socktmo
recv time: 2.016763 seconds
send time: 2.016062 seconds
$ gcc -mx32 -Wall -O2 -o socktmo socktmo.c && ./socktmo
recv time: 1.007239 seconds
send time: 1.023890 seconds
Results after patch:
$ gcc -m64 -O2 -Wall -o socktmo socktmo.c && ./socktmo
recv time: 2.010062 seconds
send time: 2.015836 seconds
$ gcc -m32 -O2 -Wall -o socktmo socktmo.c && ./socktmo
recv time: 2.013974 seconds
send time: 2.015981 seconds
$ gcc -mx32 -O2 -Wall -o socktmo socktmo.c && ./socktmo
recv time: 2.030257 seconds
send time: 2.013383 seconds
#include <stdio.h>
#include <stdlib.h>
#include <sys/socket.h>
#include <sys/types.h>
#include <sys/time.h>
void checkrc(char *str, int rc)
{
if (rc >= 0)
return;
perror(str);
exit(1);
}
static char buf[1024];
int main(int argc, char **argv)
{
int rc;
int socks[2];
struct timeval tv;
struct timeval start, end, delta;
rc = socketpair(AF_UNIX, SOCK_STREAM, 0, socks);
checkrc("socketpair", rc);
/* set timeout to 1.999999 seconds */
tv.tv_sec = 1;
tv.tv_usec = 999999;
rc = setsockopt(socks[0], SOL_SOCKET, SO_RCVTIMEO, &tv, sizeof tv);
rc = setsockopt(socks[0], SOL_SOCKET, SO_SNDTIMEO, &tv, sizeof tv);
checkrc("setsockopt", rc);
/* measure actual receive timeout */
gettimeofday(&start, NULL);
rc = recv(socks[0], buf, sizeof buf, 0);
gettimeofday(&end, NULL);
timersub(&end, &start, &delta);
printf("recv time: %ld.%06ld seconds\n",
(long)delta.tv_sec, (long)delta.tv_usec);
/* fill send buffer */
do {
rc = send(socks[0], buf, sizeof buf, 0);
} while (rc > 0);
/* measure actual send timeout */
gettimeofday(&start, NULL);
rc = send(socks[0], buf, sizeof buf, 0);
gettimeofday(&end, NULL);
timersub(&end, &start, &delta);
printf("send time: %ld.%06ld seconds\n",
(long)delta.tv_sec, (long)delta.tv_usec);
exit(0);
}
Fixes: 515c7af85ed9 ("x32: Use compat shims for {g,s}etsockopt")
Reported-by: Gopal RajagopalSai <[email protected]>
Signed-off-by: Lance Richardson <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This implements a misc character device named "qrtr-tun" for the purpose
of allowing user space applications to implement endpoints in the qrtr
network.
This allows more advanced (and dynamic) testing of the qrtr code as well
as opens up the ability of tunneling qrtr over a network or USB link.
Signed-off-by: Bjorn Andersson <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
RFC 6458 Section 8.1.16 says that setting MAXSEG as 0 means that the user
is not limiting it, and not that it should set to the *current* maximum,
as we are doing.
This patch thus allow setting it as 0, effectively removing the user
limit.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
When setting SCTP_MAXSEG sock option, it should consider which kind of
data chunk is being used if the asoc is already available, so that the
limit better reflect reality.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
sctp_sendmsg() could trigger PMTU updates even when PMTU_DISABLED was
set, as pmtu_pending could be set unconditionally during icmp handling
if the socket was in use by the application.
This patch fixes it by checking for PMTU_DISABLED when handling such
deferred updates.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
sctp_transport_route currently is very similar to sctp_transport_pmtu plus
a few other bits.
This patch reuses sctp_transport_pmtu in sctp_transport_route and removes
the duplicated code.
Also, as all calls to sctp_transport_route were forcing the dst release
before calling it, let's just include such release too.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
We are now keeping the MTU information synced between asoc, transport
and dst, which makes the check at sctp_packet_config() not needed
anymore. As it was the sole caller to this function, lets remove it.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Which makes sure that the MTU respects the minimum value of
SCTP_DEFAULT_MINSEGMENT and that it is correctly aligned.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
No need for this helper.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
and avoid the open-coded versions of it.
Now sctp_datamsg_from_user can just re-use asoc->frag_point as it will
always be updated.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
When given a MTU, this function calculates how much payload we can carry
on it. Without a MTU, it calculates the amount of header overhead we
have.
So that when we have extra overhead, like the one added for IP options
on SELinux patches, it is easier to handle it.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
All changes to asoc PMTU should now go through this wrapper, making it
easier to track them and to do other actions upon it.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
As noticed by Xin Long, the if() here is always true as PMTU can never
be 0.
Reported-by: Xin Long <[email protected]>
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
There was only one case that sctp_assoc_add_peer couldn't handle, which
is when SPP_PMTUD_DISABLE is set and pathmtu not initialized.
So add this situation to sctp_transport_route and reuse what was
already in there.
Signed-off-by: Marcelo Ricardo Leitner <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
In tcp_select_initial_window(), we only set rcv_wnd to
tcp_default_init_rwnd() if current mss > (1 << wscale). Otherwise,
rcv_wnd is kept at the full receive space of the socket which is a
value way larger than tcp_default_init_rwnd().
With larger initial rcv_wnd value, receive buffer autotuning logic
takes longer to kick in and increase the receive buffer.
In a TCP throughput test where receiver has rmem[2] set to 125MB
(wscale is 11), we see the connection gets recvbuf limited at the
beginning of the connection and gets less throughput overall.
Signed-off-by: Wei Wang <[email protected]>
Acked-by: Eric Dumazet <[email protected]>
Acked-by: Soheil Hassas Yeganeh <[email protected]>
Acked-by: Yuchung Cheng <[email protected]>
Acked-by: Neal Cardwell <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|