kernel/linux.git/include/net/tcp.h, branch v4.4.197

tcp: fix tcp_rtx_queue_tail in case of empty retransmit queue

2019-09-06T08:18:15+00:00

Commit 8c3088f895a0 ("tcp: be more careful in tcp_fragment()") triggers following stack trace: [25244.848046] kernel BUG at ./include/linux/skbuff.h:1406! [25244.859335] RIP: 0010:skb_queue_prev+0x9/0xc [25244.888167] Call Trace: [25244.889182] [25244.890001] tcp_fragment+0x9c/0x2cf [25244.891295] tcp_write_xmit+0x68f/0x988 [25244.892732] __tcp_push_pending_frames+0x3b/0xa0 [25244.894347] tcp_data_snd_check+0x2a/0xc8 [25244.895775] tcp_rcv_established+0x2a8/0x30d [25244.897282] tcp_v4_do_rcv+0xb2/0x158 [25244.898666] tcp_v4_rcv+0x692/0x956 [25244.899959] ip_local_deliver_finish+0xeb/0x169 [25244.901547] __netif_receive_skb_core+0x51c/0x582 [25244.903193] ? inet_gro_receive+0x239/0x247 [25244.904756] netif_receive_skb_internal+0xab/0xc6 [25244.906395] napi_gro_receive+0x8a/0xc0 [25244.907760] receive_buf+0x9a1/0x9cd [25244.909160] ? load_balance+0x17a/0x7b7 [25244.910536] ? vring_unmap_one+0x18/0x61 [25244.911932] ? detach_buf+0x60/0xfa [25244.913234] virtnet_poll+0x128/0x1e1 [25244.914607] net_rx_action+0x12a/0x2b1 [25244.915953] __do_softirq+0x11c/0x26b [25244.917269] ? handle_irq_event+0x44/0x56 [25244.918695] irq_exit+0x61/0xa0 [25244.919947] do_IRQ+0x9d/0xbb [25244.921065] common_interrupt+0x85/0x85 [25244.922479] tcp_rtx_queue_tail() (called by tcp_fragment()) can call tcp_write_queue_prev() on the first packet in the queue, which will trigger the BUG in tcp_write_queue_prev(), because there is no previous packet. This happens when the retransmit queue is empty, for example in case of a zero window. Commit 8c3088f895a0 ("tcp: be more careful in tcp_fragment()") was not a simple cherry-pick of the original one from master (b617158dc096) because there is a specific TCP rtx queue only since v4.15. For more details, please see the commit message of b617158dc096 ("tcp: be more careful in tcp_fragment()"). The BUG() is hit due to the specific code added to versions older than v4.15. The comment in skb_queue_prev() (include/linux/skbuff.h:1406), just before the BUG_ON() somehow suggests to add a check before using it, what Tim did. In master, this code path causing the issue will not be taken because the implementation of tcp_rtx_queue_tail() is different: tcp_fragment() → tcp_rtx_queue_tail() → tcp_write_queue_prev() → skb_queue_prev() → BUG_ON() Fixes: 8c3088f895a0 ("tcp: be more careful in tcp_fragment()") Signed-off-by: Tim Froidcoeur Signed-off-by: Matthieu Baerts Reviewed-by: Christoph Paasch Signed-off-by: Sasha Levin

tcp: be more careful in tcp_fragment()

2019-08-11T10:20:43+00:00

[ Upstream commit b617158dc096709d8600c53b6052144d12b89fab ] Some applications set tiny SO_SNDBUF values and expect TCP to just work. Recent patches to address CVE-2019-11478 broke them in case of losses, since retransmits might be prevented. We should allow these flows to make progress. This patch allows the first and last skb in retransmit queue to be split even if memory limits are hit. It also adds the some room due to the fact that tcp_sendmsg() and tcp_sendpage() might overshoot sk_wmem_queued by about one full TSO skb (64KB size). Note this allowance was already present in stable backports for kernels < 4.15 Note for < 4.15 backports : tcp_rtx_queue_tail() will probably look like : static inline struct sk_buff *tcp_rtx_queue_tail(const struct sock *sk) { struct sk_buff *skb = tcp_send_head(sk); return skb ? tcp_write_queue_prev(sk, skb) : tcp_write_queue_tail(sk); } Fixes: f070ef2ac667 ("tcp: tcp_fragment() should apply sane memory limits") Signed-off-by: Eric Dumazet Reported-by: Andrew Prout Tested-by: Andrew Prout Tested-by: Jonathan Lemon Tested-by: Michal Kubecek Acked-by: Neal Cardwell Acked-by: Yuchung Cheng Acked-by: Christoph Paasch Cc: Jonathan Looney Signed-off-by: David S. Miller Signed-off-by: Sasha Levin

tcp: reset sk_send_head in tcp_write_queue_purge

2019-08-04T07:35:01+00:00

[ Upstream commit dbbf2d1e4077bab0c65ece2765d3fc69cf7d610f ] tcp_write_queue_purge clears all the SKBs in the write queue but does not reset the sk_send_head. As a result, we can have a NULL pointer dereference anywhere that we use tcp_send_head instead of the tcp_write_queue_tail. For example, after a27fd7a8ed38 (tcp: purge write queue upon RST), we can purge the write queue on RST. Prior to 75c119afe14f (tcp: implement rb-tree based retransmit queue), tcp_push will only check tcp_send_head and then accesses tcp_write_queue_tail to send the actual SKB. As a result, it will dereference a NULL pointer. This has been reported twice for 4.14 where we don't have 75c119afe14f: By Timofey Titovets: [ 422.081094] BUG: unable to handle kernel NULL pointer dereference at 0000000000000038 [ 422.081254] IP: tcp_push+0x42/0x110 [ 422.081314] PGD 0 P4D 0 [ 422.081364] Oops: 0002 [#1] SMP PTI By Yongjian Xu: BUG: unable to handle kernel NULL pointer dereference at 0000000000000038 IP: tcp_push+0x48/0x120 PGD 80000007ff77b067 P4D 80000007ff77b067 PUD 7fd989067 PMD 0 Oops: 0002 [#18] SMP PTI Modules linked in: tcp_diag inet_diag tcp_bbr sch_fq iTCO_wdt iTCO_vendor_support pcspkr ixgbe mdio i2c_i801 lpc_ich joydev input_leds shpchp e1000e igb dca ptp pps_core hwmon mei_me mei ipmi_si ipmi_msghandler sg ses scsi_transport_sas enclosure ext4 jbd2 mbcache sd_mod ahci libahci megaraid_sas wmi ast ttm dm_mirror dm_region_hash dm_log dm_mod dax CPU: 6 PID: 14156 Comm: [ET_NET 6] Tainted: G D 4.14.26-1.el6.x86_64 #1 Hardware name: LENOVO ThinkServer RD440 /ThinkServer RD440, BIOS A0TS80A 09/22/2014 task: ffff8807d78d8140 task.stack: ffffc9000e944000 RIP: 0010:tcp_push+0x48/0x120 RSP: 0018:ffffc9000e947a88 EFLAGS: 00010246 RAX: 00000000000005b4 RBX: ffff880f7cce9c00 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000040 RDI: ffff8807d00f5000 RBP: ffffc9000e947aa8 R08: 0000000000001c84 R09: 0000000000000000 R10: ffff8807d00f5158 R11: 0000000000000000 R12: ffff8807d00f5000 R13: 0000000000000020 R14: 00000000000256d4 R15: 0000000000000000 FS: 00007f5916de9700(0000) GS:ffff88107fd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000038 CR3: 00000007f8226004 CR4: 00000000001606e0 Call Trace: tcp_sendmsg_locked+0x33d/0xe50 tcp_sendmsg+0x37/0x60 inet_sendmsg+0x39/0xc0 sock_sendmsg+0x49/0x60 sock_write_iter+0xb6/0x100 do_iter_readv_writev+0xec/0x130 ? rw_verify_area+0x49/0xb0 do_iter_write+0x97/0xd0 vfs_writev+0x7e/0xe0 ? __wake_up_common_lock+0x80/0xa0 ? __fget_light+0x2c/0x70 ? __do_page_fault+0x1e7/0x530 do_writev+0x60/0xf0 ? inet_shutdown+0xac/0x110 SyS_writev+0x10/0x20 do_syscall_64+0x6f/0x140 ? prepare_exit_to_usermode+0x8b/0xa0 entry_SYSCALL_64_after_hwframe+0x3d/0xa2 RIP: 0033:0x3135ce0c57 RSP: 002b:00007f5916de4b00 EFLAGS: 00000293 ORIG_RAX: 0000000000000014 RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 0000003135ce0c57 RDX: 0000000000000002 RSI: 00007f5916de4b90 RDI: 000000000000606f RBP: 0000000000000000 R08: 0000000000000000 R09: 00007f5916de8c38 R10: 0000000000000000 R11: 0000000000000293 R12: 00000000000464cc R13: 00007f5916de8c30 R14: 00007f58d8bef080 R15: 0000000000000002 Code: 48 8b 97 60 01 00 00 4c 8d 97 58 01 00 00 41 b9 00 00 00 00 41 89 f3 4c 39 d2 49 0f 44 d1 41 81 e3 00 80 00 00 0f 85 b0 00 00 00 <80> 4a 38 08 44 8b 8f 74 06 00 00 44 89 8f 7c 06 00 00 83 e6 01 RIP: tcp_push+0x48/0x120 RSP: ffffc9000e947a88 CR2: 0000000000000038 ---[ end trace 8d545c2e93515549 ]--- There is other scenario which found in stable 4.4: Allocated: [] __alloc_skb+0xe6/0x600 net/core/skbuff.c:218 [] alloc_skb_fclone include/linux/skbuff.h:856 [inline] [] sk_stream_alloc_skb+0xa3/0x5d0 net/ipv4/tcp.c:833 [] tcp_sendmsg+0xd34/0x2b00 net/ipv4/tcp.c:1178 [] inet_sendmsg+0x203/0x4d0 net/ipv4/af_inet.c:755 Freed: [] __kfree_skb+0x1d/0x20 net/core/skbuff.c:676 [] sk_wmem_free_skb include/net/sock.h:1447 [inline] [] tcp_write_queue_purge include/net/tcp.h:1460 [inline] [] tcp_connect_init net/ipv4/tcp_output.c:3122 [inline] [] tcp_connect+0xb24/0x30c0 net/ipv4/tcp_output.c:3261 [] tcp_v4_connect+0xf31/0x1890 net/ipv4/tcp_ipv4.c:246 BUG: KASAN: use-after-free in tcp_skb_pcount include/net/tcp.h:796 [inline] BUG: KASAN: use-after-free in tcp_init_tso_segs net/ipv4/tcp_output.c:1619 [inline] BUG: KASAN: use-after-free in tcp_write_xmit+0x3fc2/0x4cb0 net/ipv4/tcp_output.c:2056 [] kasan_report.cold.7+0x175/0x2f7 mm/kasan/report.c:408 [] __asan_report_load2_noabort+0x14/0x20 mm/kasan/report.c:427 [] tcp_skb_pcount include/net/tcp.h:796 [inline] [] tcp_init_tso_segs net/ipv4/tcp_output.c:1619 [inline] [] tcp_write_xmit+0x3fc2/0x4cb0 net/ipv4/tcp_output.c:2056 [] __tcp_push_pending_frames+0xa0/0x290 net/ipv4/tcp_output.c:2307 stable 4.4 and stable 4.9 don't have the commit abb4a8b870b5 ("tcp: purge write queue upon RST") which is referred in dbbf2d1e4077, in tcp_connect_init, it calls tcp_write_queue_purge, and does not reset sk_send_head, then UAF. stable 4.14 have the commit abb4a8b870b5 ("tcp: purge write queue upon RST"), in tcp_reset, it calls tcp_write_queue_purge(sk), and does not reset sk_send_head, then UAF. So this patch can be used to fix stable 4.4 and 4.9. Fixes: a27fd7a8ed38 (tcp: purge write queue upon RST) Reported-by: Timofey Titovets Reported-by: Yongjian Xu Signed-off-by: Eric Dumazet Signed-off-by: Soheil Hassas Yeganeh Tested-by: Yongjian Xu Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman Signed-off-by: Mao Wenan Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

tcp: limit payload size of sacked skbs

2019-06-17T17:54:22+00:00

commit 3b4929f65b0d8249f19a50245cd88ed1a2f78cff upstream. Jonathan Looney reported that TCP can trigger the following crash in tcp_shifted_skb() : BUG_ON(tcp_skb_pcount(skb) < pcount); This can happen if the remote peer has advertized the smallest MSS that linux TCP accepts : 48 An skb can hold 17 fragments, and each fragment can hold 32KB on x86, or 64KB on PowerPC. This means that the 16bit witdh of TCP_SKB_CB(skb)->tcp_gso_segs can overflow. Note that tcp_sendmsg() builds skbs with less than 64KB of payload, so this problem needs SACK to be enabled. SACK blocks allow TCP to coalesce multiple skbs in the retransmit queue, thus filling the 17 fragments to maximal capacity. CVE-2019-11477 -- u16 overflow of TCP_SKB_CB(skb)->tcp_gso_segs Backport notes, provided by Joao Martins v4.15 or since commit 737ff314563 ("tcp: use sequence distance to detect reordering") had switched from the packet-based FACK tracking and switched to sequence-based. v4.14 and older still have the old logic and hence on tcp_skb_shift_data() needs to retain its original logic and have @fack_count in sync. In other words, we keep the increment of pcount with tcp_skb_pcount(skb) to later used that to update fack_count. To make it more explicit we track the new skb that gets incremented to pcount in @next_pcount, and we get to avoid the constant invocation of tcp_skb_pcount(skb) all together. Fixes: 832d11c5cd07 ("tcp: Try to restore large SKBs while SACK processing") Signed-off-by: Eric Dumazet Reported-by: Jonathan Looney Acked-by: Neal Cardwell Reviewed-by: Tyler Hicks Cc: Yuchung Cheng Cc: Bruce Curtis Cc: Jonathan Lemon Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

tcp: clear icsk_backoff in tcp_write_queue_purge()

2019-02-23T08:05:14+00:00

[ Upstream commit 04c03114be82194d4a4858d41dba8e286ad1787c ] soukjin bae reported a crash in tcp_v4_err() handling ICMP_DEST_UNREACH after tcp_write_queue_head(sk) returned a NULL pointer. Current logic should have prevented this : if (seq != tp->snd_una || !icsk->icsk_retransmits || !icsk->icsk_backoff || fastopen) break; Problem is the write queue might have been purged and icsk_backoff has not been cleared. Signed-off-by: Eric Dumazet Reported-by: soukjin bae Acked-by: Neal Cardwell Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

tcp: use an RB tree for ooo receive queue

2018-10-13T07:11:34+00:00

[ Upstream commit 9f5afeae51526b3ad7b7cb21ee8b145ce6ea7a7a ] Over the years, TCP BDP has increased by several orders of magnitude, and some people are considering to reach the 2 Gbytes limit. Even with current window scale limit of 14, ~1 Gbytes maps to ~740,000 MSS. In presence of packet losses (or reorders), TCP stores incoming packets into an out of order queue, and number of skbs sitting there waiting for the missing packets to be received can be in the 10^5 range. Most packets are appended to the tail of this queue, and when packets can finally be transferred to receive queue, we scan the queue from its head. However, in presence of heavy losses, we might have to find an arbitrary point in this queue, involving a linear scan for every incoming packet, throwing away cpu caches. This patch converts it to a RB tree, to get bounded latencies. Yaogong wrote a preliminary patch about 2 years ago. Eric did the rebase, added ofo_last_skb cache, polishing and tests. Tested with network dropping between 1 and 10 % packets, with good success (about 30 % increase of throughput in stress tests) Next step would be to also use an RB tree for the write queue at sender side ;) Signed-off-by: Yaogong Wang Signed-off-by: Eric Dumazet Cc: Yuchung Cheng Cc: Neal Cardwell Cc: Ilpo Järvinen Acked-By: Ilpo Järvinen Signed-off-by: David S. Miller Signed-off-by: Mao Wenan Signed-off-by: Greg Kroah-Hartman

tcp: remove DELAYED ACK events in DCTCP

2018-08-24T11:26:59+00:00

[ Upstream commit a69258f7aa2623e0930212f09c586fd06674ad79 ] After fixing the way DCTCP tracking delayed ACKs, the delayed-ACK related callbacks are no longer needed Signed-off-by: Yuchung Cheng Signed-off-by: Eric Dumazet Acked-by: Neal Cardwell Acked-by: Lawrence Brakmo Signed-off-by: David S. Miller Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

tcp: add max_quickacks param to tcp_incr_quickack and tcp_enter_quickack_mode

2018-08-06T14:24:41+00:00

[ Upstream commit 9a9c9b51e54618861420093ae6e9b50a961914c5 ] We want to add finer control of the number of ACK packets sent after ECN events. This patch is not changing current behavior, it only enables following change. Signed-off-by: Eric Dumazet Acked-by: Soheil Hassas Yeganeh Acked-by: Neal Cardwell Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

tcp: do not delay ACK in DCTCP upon CE status change

2018-07-28T05:45:02+00:00

[ Upstream commit a0496ef2c23b3b180902dd185d0d63ccbc624cf8 ] Per DCTCP RFC8257 (Section 3.2) the ACK reflecting the CE status change has to be sent immediately so the sender can respond quickly: """ When receiving packets, the CE codepoint MUST be processed as follows: 1. If the CE codepoint is set and DCTCP.CE is false, set DCTCP.CE to true and send an immediate ACK. 2. If the CE codepoint is not set and DCTCP.CE is true, set DCTCP.CE to false and send an immediate ACK. """ Previously DCTCP implementation may continue to delay the ACK. This patch fixes that to implement the RFC by forcing an immediate ACK. Tested with this packetdrill script provided by Larry Brakmo 0.000 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 0.000 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0 0.000 setsockopt(3, SOL_TCP, TCP_CONGESTION, "dctcp", 5) = 0 0.000 bind(3, ..., ...) = 0 0.000 listen(3, 1) = 0 0.100 < [ect0] SEW 0:0(0) win 32792 0.100 > SE. 0:0(0) ack 1 0.110 < [ect0] . 1:1(0) ack 1 win 257 0.200 accept(3, ..., ...) = 4 +0 setsockopt(4, SOL_SOCKET, SO_DEBUG, [1], 4) = 0 0.200 < [ect0] . 1:1001(1000) ack 1 win 257 0.200 > [ect01] . 1:1(0) ack 1001 0.200 write(4, ..., 1) = 1 0.200 > [ect01] P. 1:2(1) ack 1001 0.200 < [ect0] . 1001:2001(1000) ack 2 win 257 +0.005 < [ce] . 2001:3001(1000) ack 2 win 257 +0.000 > [ect01] . 2:2(0) ack 2001 // Previously the ACK below would be delayed by 40ms +0.000 > [ect01] E. 2:2(0) ack 3001 +0.500 < F. 9501:9501(0) ack 4 win 257 Signed-off-by: Yuchung Cheng Acked-by: Neal Cardwell Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

tcp: do not cancel delay-AcK on DCTCP special ACK

2018-07-28T05:45:02+00:00

[ Upstream commit 27cde44a259c380a3c09066fc4b42de7dde9b1ad ] Currently when a DCTCP receiver delays an ACK and receive a data packet with a different CE mark from the previous one's, it sends two immediate ACKs acking previous and latest sequences respectly (for ECN accounting). Previously sending the first ACK may mark off the delayed ACK timer (tcp_event_ack_sent). This may subsequently prevent sending the second ACK to acknowledge the latest sequence (tcp_ack_snd_check). The culprit is that tcp_send_ack() assumes it always acknowleges the latest sequence, which is not true for the first special ACK. The fix is to not make the assumption in tcp_send_ack and check the actual ack sequence before cancelling the delayed ACK. Further it's safer to pass the ack sequence number as a local variable into tcp_send_ack routine, instead of intercepting tp->rcv_nxt to avoid future bugs like this. Reported-by: Neal Cardwell Signed-off-by: Yuchung Cheng Acked-by: Neal Cardwell Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman