Jakub Łągwa 338b915008 While working with version 7.10 I discovered a problem in TCP stack that could be observed on high network load. Generally speaking, the problem is that RST flag is set in unnecessary case, in which between loss of some TCP packet and its proper retransmission, another packets had been successfully sent. The scenario is as follows: NuttX did not receive ACK for some sent packet, so it has been probably lost somewhere. But before its retransmission starts, NuttX is correctly issuing next TCP packets, with sequence numbers increasing properly. When the retransmission of previously lost packet finally succeeds, tcp_input receives the accumulated ACK value, which acknowledges also the packets sent in the meantime (i.e. between unsuccessful sending of lost packet and its proper retransmission). However, variable unackseq is still set to conn->isn + conn->sent, which is truth only if no further packets transmission occurred in the meantime. Because of incorrect (in such specific case) unackseq value, few lines further condition if (ackseq <= unackseq)is not met, and, as a result, we are going to reset label.

2016-06-20 06:55:29 -06:00

..

Kconfig

Rename CONFIG_DEBUG to CONFIG_DEBUG_FEATURES

2016-06-11 14:14:08 -06:00

Make.defs

Rename CONFIG_DEBUG to CONFIG_DEBUG_FEATURES

2016-06-11 14:14:08 -06:00

tcp_accept.c

Replace all occurrences of vdbg with vinfo

2016-06-11 11:59:51 -06:00

tcp_appsend.c

Replace all occurrences of vdbg with vinfo

2016-06-11 11:59:51 -06:00

tcp_backlog.c

net/: Change some *err() message to *info() messages if what was a *dbg() message does not indicate and error condition.

2016-06-11 17:37:21 -06:00

tcp_callback.c

Rename *lldbg to *llerr

2016-06-11 14:55:27 -06:00

tcp_conn.c

While working with version 7.10 I discovered a problem in TCP stack that could be observed on high network load. Generally speaking, the problem is that RST flag is set in unnecessary case, in which between loss of some TCP packet and its proper retransmission, another packets had been successfully sent. The scenario is as follows: NuttX did not receive ACK for some sent packet, so it has been probably lost somewhere. But before its retransmission starts, NuttX is correctly issuing next TCP packets, with sequence numbers increasing properly. When the retransmission of previously lost packet finally succeeds, tcp_input receives the accumulated ACK value, which acknowledges also the packets sent in the meantime (i.e. between unsuccessful sending of lost packet and its proper retransmission). However, variable unackseq is still set to conn->isn + conn->sent, which is truth only if no further packets transmission occurred in the meantime. Because of incorrect (in such specific case) unackseq value, few lines further condition if (ackseq <= unackseq)is not met, and, as a result, we are going to reset label.

2016-06-20 06:55:29 -06:00

tcp_devpoll.c

Fix a backward conditional test introduced with the last commit

2016-03-20 09:58:07 -06:00

tcp_finddev.c

Remove some empty code section comments

2016-02-26 07:35:55 -06:00

tcp_input.c

While working with version 7.10 I discovered a problem in TCP stack that could be observed on high network load. Generally speaking, the problem is that RST flag is set in unnecessary case, in which between loss of some TCP packet and its proper retransmission, another packets had been successfully sent. The scenario is as follows: NuttX did not receive ACK for some sent packet, so it has been probably lost somewhere. But before its retransmission starts, NuttX is correctly issuing next TCP packets, with sequence numbers increasing properly. When the retransmission of previously lost packet finally succeeds, tcp_input receives the accumulated ACK value, which acknowledges also the packets sent in the meantime (i.e. between unsuccessful sending of lost packet and its proper retransmission). However, variable unackseq is still set to conn->isn + conn->sent, which is truth only if no further packets transmission occurred in the meantime. Because of incorrect (in such specific case) unackseq value, few lines further condition if (ackseq <= unackseq)is not met, and, as a result, we are going to reset label.

2016-06-20 06:55:29 -06:00

tcp_ipselect.c

Remove some empty code section comments

2016-02-26 07:35:55 -06:00

tcp_listen.c

…

tcp_netpoll.c

Rename CONFIG_DEBUG to CONFIG_DEBUG_FEATURES

2016-06-11 14:14:08 -06:00

tcp_send_buffered.c

While working with version 7.10 I discovered a problem in TCP stack that could be observed on high network load. Generally speaking, the problem is that RST flag is set in unnecessary case, in which between loss of some TCP packet and its proper retransmission, another packets had been successfully sent. The scenario is as follows: NuttX did not receive ACK for some sent packet, so it has been probably lost somewhere. But before its retransmission starts, NuttX is correctly issuing next TCP packets, with sequence numbers increasing properly. When the retransmission of previously lost packet finally succeeds, tcp_input receives the accumulated ACK value, which acknowledges also the packets sent in the meantime (i.e. between unsuccessful sending of lost packet and its proper retransmission). However, variable unackseq is still set to conn->isn + conn->sent, which is truth only if no further packets transmission occurred in the meantime. Because of incorrect (in such specific case) unackseq value, few lines further condition if (ackseq <= unackseq)is not met, and, as a result, we are going to reset label.

2016-06-20 06:55:29 -06:00

tcp_send_unbuffered.c

net/: Change some nerr() ERRORS to nwarn() WARNINGS. Anomolous network events are not errors.

2016-06-12 07:09:37 -06:00

tcp_send.c

Replace all occurrences of vdbg with vinfo

2016-06-11 11:59:51 -06:00

tcp_seqno.c

Remove some empty code section comments

2016-02-26 07:35:55 -06:00

tcp_timer.c

Replace all occurrences of vdbg with vinfo

2016-06-11 11:59:51 -06:00

tcp_wrbuffer_dump.c

Rename CONFIG_DEBUG to CONFIG_DEBUG_FEATURES

2016-06-11 14:14:08 -06:00

tcp_wrbuffer.c

Add CONFIG_DEBUG_ERROR. Change names of *dbg() * *err()

2016-06-11 15:50:49 -06:00

tcp.h

While working with version 7.10 I discovered a problem in TCP stack that could be observed on high network load. Generally speaking, the problem is that RST flag is set in unnecessary case, in which between loss of some TCP packet and its proper retransmission, another packets had been successfully sent. The scenario is as follows: NuttX did not receive ACK for some sent packet, so it has been probably lost somewhere. But before its retransmission starts, NuttX is correctly issuing next TCP packets, with sequence numbers increasing properly. When the retransmission of previously lost packet finally succeeds, tcp_input receives the accumulated ACK value, which acknowledges also the packets sent in the meantime (i.e. between unsuccessful sending of lost packet and its proper retransmission). However, variable unackseq is still set to conn->isn + conn->sent, which is truth only if no further packets transmission occurred in the meantime. Because of incorrect (in such specific case) unackseq value, few lines further condition if (ackseq <= unackseq)is not met, and, as a result, we are going to reset label.

2016-06-20 06:55:29 -06:00