The freeing and re-allocation of algorithm are protected by cpool_mutex,
so it doesn't fix an actual use-after-free, but avoids a deserved
refcount_warn_saturate() warning.
A trivial fix for the racy behavior.
Fixes: 8c73b26315aa ("net/tcp: Prepare tcp_md5sig_pool for TCP-AO")
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Dmitry Safonov <dima@arista.com>
Tested-by: Bagas Sanjaya <bagasdotme@gmail.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This reverts commit f3f32a356c0d2379d4431364e74f101f8f075ce3.
Paolo reports that the change disables autocorking even after
the userspace sets TCP_CORK.
Fixes: f3f32a356c0d ("tcp: disable tcp_autocorking for socket when TCP_NODELAY flag is set")
Link: https://lore.kernel.org/r/0d30d5a41d3ac990573016308aaeacb40a9dc79f.camel@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Based on the tcp man page, if TCP_NODELAY is set, it disables Nagle's algorithm
and packets are sent as soon as possible. However in the `tcp_push` function
where autocorking is evaluated the `nonagle` value set by TCP_NODELAY is not
considered which can trigger unexpected corking of packets and induce delays.
For example, if two packets are generated as part of a server's reply, if the
first one is not transmitted on the wire quickly enough, the second packet can
trigger the autocorking in `tcp_push` and be delayed instead of sent as soon as
possible. It will either wait for additional packets to be coalesced or an ACK
from the client before transmitting the corked packet. This can interact badly
if the receiver has tcp delayed acks enabled, introducing 40ms extra delay in
completion times. It is not always possible to control who has delayed acks
set, but it is possible to adjust when and how autocorking is triggered.
Patch prevents autocorking if the TCP_NODELAY flag is set on the socket.
Patch has been tested using an AWS c7g.2xlarge instance with Ubuntu 22.04 and
Apache Tomcat 9.0.83 running the basic servlet below:
import java.io.IOException;
import java.io.OutputStreamWriter;
import java.io.PrintWriter;
import javax.servlet.ServletException;
import javax.servlet.http.HttpServlet;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpServletResponse;
public class HelloWorldServlet extends HttpServlet {
@Override
protected void doGet(HttpServletRequest request, HttpServletResponse response)
throws ServletException, IOException {
response.setContentType("text/html;charset=utf-8");
OutputStreamWriter osw = new OutputStreamWriter(response.getOutputStream(),"UTF-8");
String s = "a".repeat(3096);
osw.write(s,0,s.length());
osw.flush();
}
}
Load was applied using wrk2 (https://github.com/kinvolk/wrk2) from an AWS
c6i.8xlarge instance. With the current auto-corking behavior and TCP_NODELAY
set an additional 40ms latency from P99.99+ values are observed. With the
patch applied we see no occurrences of 40ms latencies. The patch has also been
tested with iperf and uperf benchmarks and no regression was observed.
# No patch with tcp_autocorking=1 and TCP_NODELAY set on all sockets
./wrk -t32 -c128 -d40s --latency -R10000 http://172.31.49.177:8080/hello/hello'
...
50.000% 0.91ms
75.000% 1.12ms
90.000% 1.46ms
99.000% 1.73ms
99.900% 1.96ms
99.990% 43.62ms <<< 40+ ms extra latency
99.999% 48.32ms
100.000% 49.34ms
# With patch
./wrk -t32 -c128 -d40s --latency -R10000 http://172.31.49.177:8080/hello/hello'
...
50.000% 0.89ms
75.000% 1.13ms
90.000% 1.44ms
99.000% 1.67ms
99.900% 1.78ms
99.990% 2.27ms <<< no 40+ ms extra latency
99.999% 3.71ms
100.000% 4.57ms
Fixes: f54b311142a9 ("tcp: auto corking")
Signed-off-by: Salvatore Dipietro <dipiets@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
syzkaller report:
kernel BUG at net/core/skbuff.c:3452!
invalid opcode: 0000 [#1] PREEMPT SMP KASAN PTI
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 6.7.0-rc4-00009-gbee0e7762ad2-dirty #135
RIP: 0010:skb_copy_and_csum_bits (net/core/skbuff.c:3452)
Call Trace:
icmp_glue_bits (net/ipv4/icmp.c:357)
__ip_append_data.isra.0 (net/ipv4/ip_output.c:1165)
ip_append_data (net/ipv4/ip_output.c:1362 net/ipv4/ip_output.c:1341)
icmp_push_reply (net/ipv4/icmp.c:370)
__icmp_send (./include/net/route.h:252 net/ipv4/icmp.c:772)
ip_fragment.constprop.0 (./include/linux/skbuff.h:1234 net/ipv4/ip_output.c:592 net/ipv4/ip_output.c:577)
__ip_finish_output (net/ipv4/ip_output.c:311 net/ipv4/ip_output.c:295)
ip_output (net/ipv4/ip_output.c:427)
__ip_queue_xmit (net/ipv4/ip_output.c:535)
__tcp_transmit_skb (net/ipv4/tcp_output.c:1462)
__tcp_retransmit_skb (net/ipv4/tcp_output.c:3387)
tcp_retransmit_skb (net/ipv4/tcp_output.c:3404)
tcp_retransmit_timer (net/ipv4/tcp_timer.c:604)
tcp_write_timer (./include/linux/spinlock.h:391 net/ipv4/tcp_timer.c:716)
The panic issue was trigered by tcp simultaneous initiation.
The initiation process is as follows:
TCP A TCP B
1. CLOSED CLOSED
2. SYN-SENT --> <SEQ=100><CTL=SYN> ...
3. SYN-RECEIVED <-- <SEQ=300><CTL=SYN> <-- SYN-SENT
4. ... <SEQ=100><CTL=SYN> --> SYN-RECEIVED
5. SYN-RECEIVED --> <SEQ=100><ACK=301><CTL=SYN,ACK> ...
// TCP B: not send challenge ack for ack limit or packet loss
// TCP A: close
tcp_close
tcp_send_fin
if (!tskb && tcp_under_memory_pressure(sk))
tskb = skb_rb_last(&sk->tcp_rtx_queue); //pick SYN_ACK packet
TCP_SKB_CB(tskb)->tcp_flags |= TCPHDR_FIN; // set FIN flag
6. FIN_WAIT_1 --> <SEQ=100><ACK=301><END_SEQ=102><CTL=SYN,FIN,ACK> ...
// TCP B: send challenge ack to SYN_FIN_ACK
7. ... <SEQ=301><ACK=101><CTL=ACK> <-- SYN-RECEIVED //challenge ack
// TCP A: <SND.UNA=101>
8. FIN_WAIT_1 --> <SEQ=101><ACK=301><END_SEQ=102><CTL=SYN,FIN,ACK> ... // retransmit panic
__tcp_retransmit_skb //skb->len=0
tcp_trim_head
len = tp->snd_una - TCP_SKB_CB(skb)->seq // len=101-100
__pskb_trim_head
skb->data_len -= len // skb->len=-1, wrap around
... ...
ip_fragment
icmp_glue_bits //BUG_ON
If we use tcp_trim_head() to remove acked SYN from packet that contains data
or other flags, skb->len will be incorrectly decremented. We can remove SYN
flag that has been acked from rtx_queue earlier than tcp_trim_head(), which
can fix the problem mentioned above.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Co-developed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com>
Link: https://lore.kernel.org/r/20231210020200.1539875-1-dongchenchen2@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
After commit 939463016b7a ("tcp: change data receiver flowlabel after one dup")
we noticed an increase of TCPACKSkippedPAWS events.
Neal Cardwell tracked the issue to tcp_disordered_ack() assumption
about remote peer TS clock.
RFC 1323 & 7323 are suggesting the following:
"timestamp clock frequency in the range 1 ms to 1 sec per tick
between 1ms and 1sec."
This has to be adjusted for 1 MHz clock frequency.
This hints at reorders of SACK packets on send side,
this might deserve a future patch.
(skb->ooo_okay is always set for pure ACK packets)
Fixes: 614e8316aa4c ("tcp: add support for usec resolution in TCP TS values")
Co-developed-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: David Morley <morleyd@google.com>
Link: https://lore.kernel.org/r/20231207181342.525181-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch is based on a detailed report and ideas from Yepeng Pan
and Christian Rossow.
ACK seq validation is currently following RFC 5961 5.2 guidelines:
The ACK value is considered acceptable only if
it is in the range of ((SND.UNA - MAX.SND.WND) <= SEG.ACK <=
SND.NXT). All incoming segments whose ACK value doesn't satisfy the
above condition MUST be discarded and an ACK sent back. It needs to
be noted that RFC 793 on page 72 (fifth check) says: "If the ACK is a
duplicate (SEG.ACK < SND.UNA), it can be ignored. If the ACK
acknowledges something not yet sent (SEG.ACK > SND.NXT) then send an
ACK, drop the segment, and return". The "ignored" above implies that
the processing of the incoming data segment continues, which means
the ACK value is treated as acceptable. This mitigation makes the
ACK check more stringent since any ACK < SND.UNA wouldn't be
accepted, instead only ACKs that are in the range ((SND.UNA -
MAX.SND.WND) <= SEG.ACK <= SND.NXT) get through.
This can be refined for new (and possibly spoofed) flows,
by not accepting ACK for bytes that were never sent.
This greatly improves TCP security at a little cost.
I added a Fixes: tag to make sure this patch will reach stable trees,
even if the 'blamed' patch was adhering to the RFC.
tp->bytes_acked was added in linux-4.2
Following packetdrill test (courtesy of Yepeng Pan) shows
the issue at hand:
0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+0 bind(3, ..., ...) = 0
+0 listen(3, 1024) = 0
// ---------------- Handshake ------------------- //
// when window scale is set to 14 the window size can be extended to
// 65535 * (2^14) = 1073725440. Linux would accept an ACK packet
// with ack number in (Server_ISN+1-1073725440. Server_ISN+1)
// ,though this ack number acknowledges some data never
// sent by the server.
+0 < S 0:0(0) win 65535 <mss 1400,nop,wscale 14>
+0 > S. 0:0(0) ack 1 <...>
+0 < . 1:1(0) ack 1 win 65535
+0 accept(3, ..., ...) = 4
// For the established connection, we send an ACK packet,
// the ack packet uses ack number 1 - 1073725300 + 2^32,
// where 2^32 is used to wrap around.
// Note: we used 1073725300 instead of 1073725440 to avoid possible
// edge cases.
// 1 - 1073725300 + 2^32 = 3221241997
// Oops, old kernels happily accept this packet.
+0 < . 1:1001(1000) ack 3221241997 win 65535
// After the kernel fix the following will be replaced by a challenge ACK,
// and prior malicious frame would be dropped.
+0 > . 1:1(0) ack 1001
Fixes: 354e4aa391ed ("tcp: RFC 5961 5.2 Blind Data Injection Attack Mitigation")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Yepeng Pan <yepeng.pan@cispa.de>
Reported-by: Christian Rossow <rossow@cispa.de>
Acked-by: Neal Cardwell <ncardwell@google.com>
Link: https://lore.kernel.org/r/20231205161841.2702925-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This extra check doesn't work for a handshake when SYN segment has
(current_key.maclen != rnext_key.maclen). It could be amended to
preserve rnext_key.maclen instead of current_key.maclen, but that
requires a lookup on listen socket.
Originally, this extra maclen check was introduced just because it was
cheap. Drop it and convert tcp_request_sock::maclen into boolean
tcp_request_sock::used_tcp_ao.
Fixes: 06b22ef29591 ("net/tcp: Wire TCP-AO to request sockets")
Signed-off-by: Dmitry Safonov <dima@arista.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
If the connection was established, don't allow adding TCP-AO keys that
don't match the peer. Currently, there are checks for ip-address
matching, but L3 index check is missing. Add it to restrict userspace
shooting itself somewhere.
Yet, nothing restricts the CAP_NET_RAW user from trying to shoot
themselves by performing setsockopt(SO_BINDTODEVICE) or
setsockopt(SO_BINDTOIFINDEX) over an established TCP-AO connection.
So, this is just "minimum effort" to potentially save someone's
debugging time, rather than a full restriction on doing weird things.
Fixes: 248411b8cb89 ("net/tcp: Wire up l3index to TCP-AO")
Signed-off-by: Dmitry Safonov <dima@arista.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Listen socket is not an established TCP connection, so
setsockopt(TCP_AO_REPAIR) doesn't have any impact.
Restrict this uAPI for listen sockets.
Fixes: faadfaba5e01 ("net/tcp: Add TCP_AO_REPAIR")
Signed-off-by: Dmitry Safonov <dima@arista.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Currently functions that pre-calculate TCP header options length use
unaligned TCP-AO header + MAC-length for skb reservation.
And the functions that actually write TCP-AO options into skb do align
the header. Nothing good can come out of this for ((maclen % 4) != 0).
Provide tcp_ao_len_aligned() helper and use it everywhere for TCP
header options space calculations.
Fixes: 1e03d32bea8e ("net/tcp: Add TCP-AO sign to outgoing packets")
Signed-off-by: Dmitry Safonov <dima@arista.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>