Gitiles
Code Review
Sign In
git01.mediatek.com
/
haproxy
/
edcf6687d68cbc2d612d9ae2dd36028ec99503d1
edcf668
[MEDIUM] extract TCP request processing from HTTP
by Willy Tarreau
· Sun Nov 30 23:15:34 2008 +0100
b025325
[MINOR] stream_sock_data_finish() should not expose fd
by Willy Tarreau
· Sun Nov 30 21:37:12 2008 +0100
42ffbf2
[CLEANUP] session.c: removed some migration left-overs in sess_establish()
by Willy Tarreau
· Sun Nov 30 21:13:54 2008 +0100
0cac36f
[MEDIUM] make the http server error function a pointer in the session
by Willy Tarreau
· Sun Nov 30 20:44:17 2008 +0100
2d3d94c
[MINOR] replace srv_close_with_err() with http_server_error()
by Willy Tarreau
· Sun Nov 30 20:20:08 2008 +0100
dded32d
[MINOR] replace client_retnclose() with stream_int_retnclose()
by Willy Tarreau
· Sun Nov 30 19:48:07 2008 +0100
81acfab
[MINOR] replace the ambiguous client_return function by stream_int_return
by Willy Tarreau
· Sun Nov 30 19:22:53 2008 +0100
a5555ec
[MINOR] call session->do_log() for logging
by Willy Tarreau
· Sun Nov 30 19:02:32 2008 +0100
55a8d0e
[CLEANUP] move the session-related functions to session.c
by Willy Tarreau
· Sun Nov 30 18:47:21 2008 +0100
fe3718a
[MAJOR] complete layer4/7 separation
by Willy Tarreau
· Sun Nov 30 18:14:12 2008 +0100
99126c3
[MEDIUM] make the stream interface control the SHUT{R,W} bits
by Willy Tarreau
· Thu Nov 27 10:30:51 2008 +0100
8bfa426
[MEDIUM] process shutw during connection attempt
by Willy Tarreau
· Thu Nov 27 09:25:45 2008 +0100
b38903c
[BUG] shutw must imply close during a connect
by Willy Tarreau
· Sun Nov 23 21:33:29 2008 +0100
f54f8bd
[MINOR] maintain a global session list in order to ease debugging
by Willy Tarreau
· Sun Nov 23 19:53:55 2008 +0100
0a5d5dd
[MEDIUM] remove stream_sock_update_data()
by Willy Tarreau
· Sun Nov 23 19:31:35 2008 +0100
74ab2ac
[MEDIUM] stream_interface: added a DISconnected state between CON/EST and CLO
by Willy Tarreau
· Sun Nov 23 17:23:07 2008 +0100
4351b3a
[MEDIUM] continue layering cleanups.
by Willy Tarreau
· Wed Nov 12 01:51:41 2008 +0100
1e62de6
[MEDIUM] add the SN_CURR_SESS flag to the session to track open sessions
by Willy Tarreau
· Tue Nov 11 20:20:02 2008 +0100
cff6411
[MAJOR] add a connection error state to the stream_interface
by Willy Tarreau
· Mon Nov 03 06:26:53 2008 +0100
efb453c
[MAJOR] migrate the connection logic to stream interface
by Willy Tarreau
· Sun Oct 26 20:49:47 2008 +0100
d7704b5
[MINOR] add an expiration flag to the stream_sock_interface
by Willy Tarreau
· Thu Sep 04 11:51:16 2008 +0200
3c6ab2e
[MEDIUM] use buffer_check_timeouts instead of stream_sock_check_timeouts()
by Willy Tarreau
· Thu Sep 04 11:19:41 2008 +0200
2eb52f0
[MINOR] add buffer_check_timeouts() to check what timeouts have fired.
by Willy Tarreau
· Thu Sep 04 09:14:08 2008 +0200
534c5a2
[OPTIM] add compiler hints in tick_is_expired()
by Willy Tarreau
· Thu Sep 04 09:00:24 2008 +0200
3537467
[MEDIUM] move QUEUE and TAR timers to stream interfaces
by Willy Tarreau
· Wed Sep 03 18:11:02 2008 +0200
a37095b
[CLEANUP] process_session: move debug outputs out of the critical loop
by Willy Tarreau
· Wed Sep 03 11:37:47 2008 +0200
4ffd51a
[MEDIUM] process_session: make use of the new buffer flags
by Willy Tarreau
· Sat Aug 30 13:36:43 2008 +0200
9a2d154
[MEDIUM] buffers: add BF_READ_ATTACHED and BF_ANA_TIMEOUT
by Willy Tarreau
· Sat Aug 30 12:31:07 2008 +0200
48adac5
[MEDIUM] stream interface: add the ->shutw method as well as in and out buffers
by Willy Tarreau
· Sat Aug 30 04:58:38 2008 +0200
e5ed406
[MAJOR] make stream sockets aware of the stream interface
by Willy Tarreau
· Sat Aug 30 03:17:31 2008 +0200
eabf313
[MINOR] change type of fdtab[]->owner to void*
by Willy Tarreau
· Fri Aug 29 23:36:51 2008 +0200
fdccded
[MEDIUM] indicate a reason for a task wakeup
by Willy Tarreau
· Fri Aug 29 18:19:04 2008 +0200
75cf17e
[OPTIM] force inlining of large functions with gcc >= 3
by Willy Tarreau
· Fri Aug 29 15:48:49 2008 +0200
4df8206
[OPTIM] reduce the number of calls to task_wakeup()
by Willy Tarreau
· Fri Aug 29 15:26:14 2008 +0200
cb65125
[OPTIM] ev_sepoll: detect newly created FDs and check them once
by Willy Tarreau
· Fri Aug 29 13:57:30 2008 +0200
21e1be8
[MINOR] do not check for BF_SHUTR when computing write timeout
by Willy Tarreau
· Fri Aug 29 11:30:14 2008 +0200
3da77c5
[MINOR] re-arrange buffer flags and rename some of them
by Willy Tarreau
· Fri Aug 29 09:58:42 2008 +0200
72b179a
[MEDIUM] reintroduce BF_HIJACK with produce_content
by Willy Tarreau
· Thu Aug 28 16:01:32 2008 +0200
36e6a41
[MINOR] only call flow analysers when their read side is connected.
by Willy Tarreau
· Thu Aug 28 11:07:58 2008 +0200
2bea3a1
[OPTIM] stream_sock_read must check for null-reads more often
by Willy Tarreau
· Thu Aug 28 09:47:43 2008 +0200
3a16b2c
[MEDIUM] split stream_sock_process_data
by Willy Tarreau
· Thu Aug 28 08:54:27 2008 +0200
f9839bd
[MAJOR] make the client side use stream_sock_process_data()
by Willy Tarreau
· Wed Aug 27 23:57:16 2008 +0200
2d21279
[MEDIUM] stream_sock_process_data moved to stream_sock.c
by Willy Tarreau
· Wed Aug 27 21:41:35 2008 +0200
8a81883
[MEDIUM] process_srv_data: ensure that we always correctly re-arm timeouts
by Willy Tarreau
· Wed Aug 27 21:10:25 2008 +0200
2ac679d
[MEDIUM] third cleanup and optimization of process_srv_data()
by Willy Tarreau
· Wed Aug 27 20:35:41 2008 +0200
8fbd3b4
[MEDIUM] second level of code cleanup for process_srv_data
by Willy Tarreau
· Wed Aug 27 20:09:19 2008 +0200
376580a
[MEDIUM] massive cleanup of process_srv()
by Willy Tarreau
· Wed Aug 27 18:52:22 2008 +0200
8b46aa0
[OPTIM] remove useless fd_set(read) upon shutdown(write)
by Willy Tarreau
· Wed Aug 27 15:19:25 2008 +0200
fa7e102
[MAJOR] rework of the server FSM
by Willy Tarreau
· Sun Oct 19 07:30:41 2008 +0200
41f40ed
[MEDIUM] make it possible for analysers to follow the whole session
by Willy Tarreau
· Thu Aug 21 10:05:00 2008 +0200
788e284
[BUG] fix harmless but wrong fd insertion sequence
by Willy Tarreau
· Tue Aug 26 13:25:39 2008 +0200
79f5fe8
[BUG] Fix empty X-Forwarded-For header name when set in defaults section
by Willy Tarreau
· Sat Aug 23 08:18:21 2008 +0200
c22b57d
[TESTS] test-fsm: 22 regression tests for state machines
by Willy Tarreau
· Sun Aug 17 19:30:03 2008 +0200
c52164a
[BUG] process_request: HTTP body analysis must return zero if missing data
by Willy Tarreau
· Sun Aug 17 19:17:57 2008 +0200
2500981
[BUG] process_cli/process_srv: don't call shutdown when already done
by Willy Tarreau
· Sun Aug 17 18:16:38 2008 +0200
ffab5b4
[MEDIUM] merge inspect_exp and txn->exp into request buffer
by Willy Tarreau
· Sun Aug 17 18:03:28 2008 +0200
c7e961e
[BUILD] fix warning in proto_tcp.c with gcc >= 4
by Willy Tarreau
· Sun Aug 17 17:13:47 2008 +0200
61eadc0
[BUG] regparm is broken on gcc < 3
by Willy Tarreau
· Sun Aug 17 17:06:37 2008 +0200
6d2889b
[OPTIM] process_cli/process_srv: reduce the number of tests
by Willy Tarreau
· Sun Aug 17 16:23:10 2008 +0200
2df28e8
[MEDIUM] session: move the analysis bit field to the buffer
by Willy Tarreau
· Sun Aug 17 15:20:19 2008 +0200
f495ddf
[MINOR] ensure the termination flags are set by process_xxx
by Willy Tarreau
· Sun Aug 17 14:38:41 2008 +0200
507385d
[MEDIUM] centralize buffer timeout checks at the top of process_session
by Willy Tarreau
· Sun Aug 17 13:04:25 2008 +0200
26ed74d
[MEDIUM] use buffer->wex instead of buffer->cex for connect timeout
by Willy Tarreau
· Sun Aug 17 12:11:14 2008 +0200
dafde43
[MAJOR] process_session: rely only on buffer flags
by Willy Tarreau
· Sun Aug 17 01:00:46 2008 +0200
e393fe2
[MEDIUM] buffers: add BF_EMPTY and BF_FULL to remove dependency on req/rep->l
by Willy Tarreau
· Sat Aug 16 22:18:07 2008 +0200
ba392ce
[CLEANUP] get rid of BF_SHUT*_PENDING
by Willy Tarreau
· Sat Aug 16 21:13:23 2008 +0200
d5382b4
[BUG] maintain_proxies must not disable backends
by Willy Tarreau
· Sat Aug 16 18:41:13 2008 +0200
a7c5276
[BUG] process_response: do not touch srv_state
by Willy Tarreau
· Sat Aug 16 18:40:18 2008 +0200
d9f4836
[BUG] buffers: remove BF_MAY_CONNECT and fix forwarding issue
by Willy Tarreau
· Sat Aug 16 16:39:26 2008 +0200
9a8c5de
[BUG] process_response must not enable the read FD
by Willy Tarreau
· Sat Aug 16 16:11:07 2008 +0200
7a52a5c
[BUG] ev_sepoll: closed file descriptors could persist in the spec list
by Willy Tarreau
· Sat Aug 16 16:06:02 2008 +0200
f853320
[MINOR] term_trace: add better instrumentations to trace the code
by Willy Tarreau
· Sat Aug 16 14:55:08 2008 +0200
1ae3a05
[MEDIUM] remove unused references to {CL|SV}_STSHUT*
by Willy Tarreau
· Sat Aug 16 10:56:30 2008 +0200
461f662
[MAJOR] clearly separate HTTP response processing from TCP server state
by Willy Tarreau
· Fri Aug 15 23:43:19 2008 +0200
cebf57e
[MAJOR] better separation of response processing and server state
by Willy Tarreau
· Fri Aug 15 18:16:37 2008 +0200
f5483bf
[MAJOR] get rid of the SV_STHEADERS state
by Willy Tarreau
· Thu Aug 14 18:35:40 2008 +0200
e46ab55
[BUG] fix recently introduced loop when client closes early
by Willy Tarreau
· Wed Aug 13 19:19:37 2008 +0200
c65a3ba
[MAJOR] completely separate HTTP and TCP states on the request path
by Willy Tarreau
· Mon Aug 11 23:42:50 2008 +0200
7f875f6
[MEDIUM] simplify and centralize request timeout cancellation and request forwarding
by Willy Tarreau
· Mon Aug 11 17:35:01 2008 +0200
adfb856
[MAJOR] get rid of SV_STANALYZE (step 2)
by Willy Tarreau
· Mon Aug 11 15:24:42 2008 +0200
67f0eea
[MAJOR] kill CL_STINSPECT and CL_STHEADERS (step 1)
by Willy Tarreau
· Sun Aug 10 22:55:22 2008 +0200
697bbb0
[PATCH] appsessions: cleanup DEBUG_HASH and initialize request_counter
by Aleksandar Lazic
· Wed Aug 13 19:57:02 2008 +0200
9f1f24b
[BUG] client timeout incorrectly rearmed while waiting for server
by Willy Tarreau
· Mon Aug 11 11:20:03 2008 +0200
ce09c52
[BUG] server timeout was not considered in some circumstances
by Willy Tarreau
· Mon Aug 11 10:35:07 2008 +0200
fb0528b
[BUG] fix segfault with url_param + check_post
by Willy Tarreau
· Mon Aug 11 00:21:56 2008 +0200
718f0ef
[MEDIUM] process_cli: don't rely at all on server state
by Willy Tarreau
· Sun Aug 10 16:21:32 2008 +0200
dc0a6a0
[MEDIUM] process_srv: don't rely at all on client state
by Willy Tarreau
· Sun Aug 03 20:38:13 2008 +0200
6468d92
[MEDIUM] process_srv: rely on buffer flags for client shutdown
by Willy Tarreau
· Sun Aug 03 19:15:35 2008 +0200
89edf5e
[MEDIUM] buffers: ensure buffer_shut* are properly called upon shutdowns
by Willy Tarreau
· Sun Aug 03 17:25:14 2008 +0200
48d63db
[MEDIUM] memory: update pool_free2() to support NULL pointers
by Willy Tarreau
· Sun Aug 03 17:41:33 2008 +0200
a534fea
[CLEANUP] remove 65 useless NULL checks before free
by Willy Tarreau
· Sun Aug 03 12:19:50 2008 +0200
af72a1d
[MINOR] permit renaming of x-forwarded-for header
by Ross West
· Sun Aug 03 10:51:45 2008 +0200
dd64f8d
[MEDIUM] acl: when possible, report the name and requirements of ACLs in warnings
by Willy Tarreau
· Sun Jul 27 22:02:32 2008 +0200
0ceba5a
[MEDIUM] acl: set types on all currently known ACL verbs
by Willy Tarreau
· Fri Jul 25 19:31:03 2008 +0200
a980263
[MEDIUM] acl: enforce ACL type checking
by Willy Tarreau
· Fri Jul 25 19:13:19 2008 +0200
b6fb420
[MINOR] acl: add the "wait_end" acl verb
by Willy Tarreau
· Sun Jul 20 11:18:28 2008 +0200
58393e1
[MEDIUM] acl: get rid of dummy values in always_true/always_false
by Willy Tarreau
· Sun Jul 20 10:39:22 2008 +0200
a79534f
[MEDIUM] acl: permit fetch() functions to set the result themselves
by Willy Tarreau
· Sun Jul 20 10:13:37 2008 +0200
c631770
[MINOR] acl: add REQ_CONTENT to the list of default acls
by Willy Tarreau
· Sun Jul 20 09:29:50 2008 +0200
6264477
[DOC] document the new "tcp-request" keyword and associated ACLs
by Willy Tarreau
· Wed Jul 16 18:36:06 2008 +0200
Next »