Willy Tarreau 3c5c066d66 MEDIUM: stream: only loop on flags relevant to the analysers ...

In process_stream() we detect a number of conditions to decide to loop
back to the analysers. Some of them are excessive in that they perform
a strict comparison instead of filtering on the flags relevant to the
analysers as is done at other places, resulting in excess wakeups. One
of the effect is that after a successful WRITE_PARTIAL, a second send is
not possible, resulting in the loss of WRITE_PARTIAL, causing another
wakeup! Let's apply the same mask and verify the flags correctly.

2019-06-06 16:36:19 +02:00

..

51d.c

…

acl.c

…

action.c

…

activity.c

…

applet.c

…

arg.c

…

auth.c

…

backend.c

MEDIUM: connections: Introduce a handshake pseudo-XPRT.

2019-06-05 18:03:38 +02:00

base64.c

…

buffer.c

…

cache.c

…

cfgparse-global.c

…

cfgparse-listen.c

…

cfgparse.c

MINOR: server: really increase the pool-purge-delay default to 5 seconds

2019-06-06 16:25:55 +02:00

channel.c

…

checks.c

MEDIUM: connections: Introduce a handshake pseudo-XPRT.

2019-06-05 18:03:38 +02:00

chunk.c

…

cli.c

…

compression.c

…

connection.c

MEDIUM: connections: Remove CONN_FL_SOCK*

2019-06-05 18:03:38 +02:00

da.c

…

debug.c

…

dict.c

…

dns.c

…

ev_epoll.c

…

ev_evports.c

…

ev_kqueue.c

…

ev_poll.c

…

ev_select.c

…

fd.c

…

filters.c

…

flt_http_comp.c

…

flt_spoe.c

MEDIUM: stream-int: introduce a new state SI_ST_RDY

2019-06-06 16:36:19 +02:00

flt_trace.c

BUG/MINOR: flt_trace/htx: Only apply the random forwarding on the message body.

2019-06-05 10:12:11 +02:00

freq_ctr.c

…

frontend.c

…

h1.c

…

h2.c

…

haproxy.c

…

hash.c

…

hathreads.c

…

hdr_idx.c

…

hlua_fcn.c

…

hlua.c

…

hpack-dec.c

…

hpack-enc.c

…

hpack-huff.c

…

hpack-tbl.c

…

http_acl.c

…

http_act.c

…

http_conv.c

…

http_fetch.c

…

http_htx.c

…

http_msg.c

…

http_rules.c

…

http.c

…

htx.c

…

i386-linux-vsys.c

…

lb_chash.c

…

lb_fas.c

…

lb_fwlc.c

…

lb_fwrr.c

…

lb_map.c

…

listener.c

…

log.c

…

lru.c

…

mailers.c

…

map.c

…

memory.c

…

mux_h1.c

BUG/MINOR: mux-h1: Don't send more data than expected

2019-06-05 10:12:11 +02:00

mux_h2.c

…

mux_pt.c

…

mworker-prog.c

…

mworker.c

…

namespace.c

…

pattern.c

…

payload.c

…

peers.c

BUG/MINOR: peers: Wrong server name parsing.

2019-06-06 16:06:00 +02:00

pipe.c

…

proto_http.c

…

proto_htx.c

…

proto_sockpair.c

MEDIUM: connections: Remove CONN_FL_SOCK*

2019-06-05 18:03:38 +02:00

proto_tcp.c

MEDIUM: connections: Remove CONN_FL_SOCK*

2019-06-05 18:03:38 +02:00

proto_udp.c

…

proto_uxst.c

MEDIUM: connections: Remove CONN_FL_SOCK*

2019-06-05 18:03:38 +02:00

protocol.c

…

proxy.c

…

queue.c

…

raw_sock.c

MEDIUM: connections: Remove CONN_FL_SOCK*

2019-06-05 18:03:38 +02:00

regex.c

…

sample.c

…

server.c

MINOR: server: really increase the pool-purge-delay default to 5 seconds

2019-06-06 16:25:55 +02:00

session.c

MEDIUM: connections: Introduce a handshake pseudo-XPRT.

2019-06-05 18:03:38 +02:00

sha1.c

…

shctx.c

…

signal.c

…

ssl_sock.c

BUG/MEDIUM: ssl: Don't forget to initialize ctx->send_recv and ctx->recv_wait.

2019-06-06 13:21:23 +02:00

standard.c

…

stats.c

…

stick_table.c

…

stream_interface.c

MEDIUM: stream: rearrange the events to remove the loop

2019-06-06 16:36:19 +02:00

stream.c

MEDIUM: stream: only loop on flags relevant to the analysers

2019-06-06 16:36:19 +02:00

task.c

…

tcp_rules.c

MEDIUM: connections: Introduce a handshake pseudo-XPRT.

2019-06-05 18:03:38 +02:00

time.c

…

trace.c

…

uri_auth.c

…

vars.c

…

version.c

…

wdt.c

…

wurfl.c

…

xprt_handshake.c

MEDIUM: connections: Introduce a handshake pseudo-XPRT.

2019-06-05 18:03:38 +02:00

xxhash.c

…