Author: John Buttery <john@io.com>

[ircu2.10.12-pk.git] / ChangeLog

2002-02-19 John Buttery <john@io.com>

        * ircd/ircd.c: Updated "No such file" error message.
        
2002-02-18 Joseph Bongaarts <foxxe@wtfs.net>

        * ircd/m_kill.c: Changed m_kill() to do_kill() because its not
        a message handler, and some general cleanups and bug fixes. 

        * include/ircd_policy.h: Added HEAD_IN_SAND_KILLWHO for hiding
        kill source.
        
2002-02-16  Tim Vogelsang <net@astrolink.org>

        * ircd/m_kill.c: added a new static function, m_kill, which
        performs the actual kill.
        
2002-02-14 Joseph Bongaarts <foxxe@wtfs.net>

        * Added support for LIST STOP
        
2002-02-13 Joseph Bongaarts <foxxe@wtfs.net>

        * Merged changes from u2_10_11 to main branch.
        
2002-02-08  Tim Vogelsang  <net@astrolink.org>

       * ircd/m_quit.c: don't prefix user quits with "Quit:" unless a
       reason is supplied.

2002-02-06  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/s_auth.c (read_auth_reply): left out an = in an
        assertion--shouldn't have had any impact, though

        * ircd/Makefile.in: add a hook for using ringlog; run make depend

        * tools/ringlog.c: for the heck of it, add a comment including
        rules for /etc/magic

2002-02-05  Kevin L Mitchell  <klmitch@mit.edu>

        * tools/ringlog.pl: perl script to take output from ringlog and
        pass it to addr2line to get function, file, and line number
        information

        * tools/ringlog.c: program/object to help in building function
        trace information

2002-02-04  Alex Badea  <vampire@p16.pub.ro>

        * include/ircd_features.h: added new feature MOTD_BANNER

        * ircd/ircd_features.c: added new feature MOTD_BANNER

        * ircd/motd.c (motd_signon): send a one-line banner from
        FEAT_MOTD_BANNER if it's not NULL and FEAT_NODEFAULTMOTD
        is set

        * doc/example.conf: default value for MOTD_BANNER feature

        * doc/readme.features: documented the MOTD_BANNER feature

2002-02-04  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/s_debug.c (debug_serveropts): remove deprecated CHROOTDIR
        check; added character 'A' to the server options string to
        indicate when assertion checking is enabled

2002-02-03  Kevin L Mitchell  <klmitch@mit.edu>

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/engine_kqueue.c (set_or_clear): don't generate an ET_ERROR
        event if the errno is EBADF, since the caller probably already
        knows about it and just hasn't gotten around to processing it yet

        * ircd/ircd_events.c: set the GEN_ERROR flag if an ET_ERROR event
        is generated; don't process socket_events() or socket_state() if
        an error occurred; add GEN_ERROR to list of flags in gen_flags()

        * include/ircd_events.h: define new GEN_ERROR flag; add a macro to
        clear it

2002-02-01  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/channel.c: change make_nick_user_{ip,host} to not use a
        static buffer--instead, a buffer of the right size (NUH_BUFSIZE or
        NUI_BUFSIZE--I confess they're not well-named) is allocated by the
        caller

2002-02-02  Alex Badea  <vampire@p16.pub.ro>

        * include/client.h: added user flag FLAGS_HIDDENHOST

        * include/ircd_features.h: added FEAT_HOST_HIDING and
        FEAT_HIDDEN_HOST

        * include/numeric.h: defined numeric 338 (RPL_WHOISACTUALLY)
        to report real hostnames and IPs to opers

        * include/s_user.h: exported hide_hostmask()

        * include/send.h: changed sendcmdto_channel_butserv to
        sendcmdto_channel_butserv_butone; ditto for
        sendcmdto_common_channels

        * include/struct.h: added realhost to struct User

        * include/whowas.h: added realhost to struct Whowas

        * ircd/channel.c: match bans against both real and hidden
        hostmasks; moved some calls to use sendcmdto_*_butone

        * ircd/gline.c: match glines agains real host

        * ircd/ircd_features.c: added FEAT_HOST_HIDING and
        FEAT_HIDDEN_HOST

        * ircd/m_account.c: call hide_hostmask() for possibly
        hiding the user's host

        * ircd/m_burst.c: moved some calls to use sendcmdto_*_butone

        * ircd/m_topic.c: moved some calls to use sendcmdto_*_butone

        * ircd/m_userip.c: report IP 127.0.0.1 if the user has a hidden
        host

        * ircd/m_who.c: match real hosts, if the query comes from an oper

        * ircd/m_whois.c: report real hostname and IP to opers

        * ircd/m_whowas.c: report real hostname to opers

        * ircd/s_err.c: added user mode 'x' to the list of supported user
        modes in RPL_MYINFO (004); added RPL_WHOISACTUALLY for reporting
        real hostnames to opers

        * ircd/s_misc.c: moved some calls to use sendcmdto_*_butone

        * ircd/s_serv.c: send real hostname to servers

        * ircd/s_user.c: send real hostname to servers; added processing
        of usermode 'x'; added hide_hostmask() which actually does the work
        of hiding a user's host; moved some calls to use sendcmdto_*_butone

        * ircd/send.c: changed sendcmdto_channel_butserv to
        sendcmdto_channel_butserv_butone; ditto for
        sendcmdto_common_channels

        * ircd/whocmds.c: extra letter 'x' in WHO reply if the user has
        it's host hidden

        * ircd/whowas.c: if needed, store a user's real host so we can
        report it to opers later

        * doc/readme.features: documented HOST_HIDING and HIDDEN_HOST
        features

        * doc/example.conf: default values for HOST_HIDING and
        HIDDEN_HOST features

2002-02-01  Tim Vogelsang  <net@astrolink.org>

        * ircd/send.c (sendwallto_group_butone): don't sent wallops to
        ordinary users

2002-01-28  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/jupe.c (jupe_activate): remove a bogus assertion

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/s_err.c: added new channel mode 'r' to list of supported
        channel modes in RPL_MYINFO (004); migrated RPL_USERIP to use
        numeric 340 instead of 307; add ERR_NEEDREGGEDNICK (477) for
        informing users why they can't join a +r channel

        * ircd/m_clearmode.c (do_clearmode): add support for MODE_REGONLY
        (+r) to do_clearmode(); note that it is *not* being added to the
        default clearmode mask!

        * ircd/channel.c: don't allow non-+r users to send messages to +r
        channels from off the channel; add support for MODE_REGONLY (+r)
        to channel_modes(); don't allow non-+r users to join +r channels
        without an invite; add support for MODE_REGONLY to the modebuf_*()
        family of functions (changes in modebuf_flush_int(),
        modebuf_mode(), and modebuf_extract()); add support for
        MODE_REGONLY to mode_parse()

        * include/supported.h (FEATURESVALUES2): added the new channel
        mode 'r' to the list of supported channel modes

        * include/numeric.h: move RPL_USERIP to 340 to avoid the 307
        conflict; add ERR_NEEDREGGEDNICK (477) for the new +r channels

        * include/channel.h: remove unused MODE_SENDTS; add new
        MODE_REGONLY

        * ircd/s_bsd.c (read_packet): remove call to timer_verify()

        * ircd/list.c: remove calls to timer_verify() from
        alloc_connection() and dealloc_connection()

        * ircd/ircd_events.c: turn off timer_verify(); remove calls to it
        from timer_run()

2002-01-27  Kevin L Mitchell  <klmitch@mit.edu>

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/ircd_events.c (timer_run): why did I ever use a next
        pointer when the algorithm guarantees that the head pointer will
        always be the next pointer?

2002-01-26  Kevin L Mitchell  <klmitch@mit.edu>

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/s_bsd.c (read_packet): call timer_verify() after adding the
        client process timer to catch any list corruption

        * ircd/list.c: surround alloc_connection() and
        dealloc_connection() with calls to timer_verify()

        * ircd/ircd_events.c: add sledgehammer known as timer_verify() to
        verify the timer list's structure; call it around timer_run()

2002-01-22  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/send.c (sendcmdto_common_channels): don't send message to a
        channel that the source is a zombie on

2002-01-13  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/ircd_events.c (timer_enqueue): one more assertion--make
        sure a timer has the ACTIVE flag set before enqueueing the timer

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/list.c (dealloc_connection): assert that the process timer
        has been removed from the timer queue before dealloc_connection()
        is called

2002-01-12  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/res.c: don't accept T_A when we're looking for T_PTR

        * ircd/channel.c (modebuf_flush_int): nuke the code that would
        send a HACK DESYNCH notice on a HACK(2)--it would be far too
        chatty

        * ircd/m_away.c (user_set_away): use AWAYLEN instead of TOPICLEN

        * include/supported.h: add AWAYLEN to the list of supported
        features

        * include/ircd_defs.h: add AWAYLEN to specify the maximum length
        of an away message

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/m_mode.c (m_mode): pass extra parameter to channel_modes()

        * ircd/channel.c: pass a buflen parameter to channel_modes() for
        pbuf--we were using sizeof(pbuf), which would always be
        sizeof(char*) before; change send_channel_modes() to pass extra
        parameter to channel_modes()

        * include/channel.h: pass a buflen parameter to channel_modes()
        for pbuf

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/uping.c (uping_start): initialize some timers

        * ircd/s_bsd.c (read_packet): use new t_onqueue() macro to figure
        out when we need to re-add the process timer

        * ircd/s_auth.c (make_auth_request): initialize a timer

        * ircd/res.c (init_resolver): initialize some timers

        * ircd/list.c (alloc_connection): initialize the client process
        timer

        * ircd/ircd_events.c: add a function, timer_init(), to initialize
        a struct Timer; recast timer_add() to catch when adding a marked
        timer and not re-enqueue it--but mark it for re-enqueuing; update
        timer_del() to turn off the GEN_READD flag and to ignore reference
        counts when destroying the timer--we're using GEN_MARKED as an
        ersatz referance count; changed timer_run() to work with the new
        way of doing things; add GEN_ACTIVE and GEN_READD to gen_flags()'s
        map[]

        * ircd/ircd.c: initialize some timers

        * ircd/engine_select.c (engine_loop): initialize a timer

        * ircd/engine_poll.c (engine_loop): initialize a timer

        * ircd/engine_kqueue.c (engine_loop): initialize a timer

        * ircd/engine_devpoll.c (engine_loop): initialize a timer

        * ircd/IPcheck.c (IPcheck_init): initialize a timer

        * include/ircd_events.h: add GEN_READD flag for timers to indicate
        that a timer must be readded; add t_onqueue() macro to check to
        see if a timer is on the queue (this is a hack, though); added
        timer_init() to initialize a struct Timer--we're no longer doing
        the initialization in timer_add()

2002-01-11  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/engine_devpoll.c (engine_loop): relocate an assertion to
        prevent a core bug *in* the assertion

        * doc/readme.features: document new POLLS_PER_LOOP feature; change
        documentation to reflect that OPER_SET now defaults to FALSE

        * doc/p10.html: documented the new ACCOUNT stuff

        * doc/example.conf: document new POLLS_PER_LOOP default; change
        default for OPER_SET

        * RELEASE.NOTES: changed documentation to reflect the fact that
        assertions are now enabled by default and do not cause memory
        leaks

        * ircd/res.c (make_cache): removed a bogus assertion we probably
        never caught because assertions haven't been enabled on production
        servers for any length of time before

        * ircd/engine_devpoll.c (engine_loop): ditto for POLLS_PER_DEVPOLL

        * ircd/engine_kqueue.c (engine_loop): stupid me forgot one
        instance of POLLS_PER_KQUEUE

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/s_bsd.c (client_timer_callback): only clear the
        FREEFLAG_TIMER flag when the timer is being destroyed

        * ircd/ircd_features.c: create a new feature, POLLS_PER_LOOP, and
        default it to 200; turn OPER_SET off by default

        * ircd/engine_kqueue.c: dynamically allocate and reallocate the
        array of events to obtain from the kernel

        * ircd/engine_devpoll.c: dynamically allocate and reallocate the
        array of events to obtain from the kernel

        * include/ircd_features.h: add a new feature for tuning how many
        events to get from the kernel, for engines that support that

        * ircd/Makefile.in: re-run make depend to correct dependancies

        * ircd/m_who.c: remove unneeded inclusion of list.h

        * ircd/ircd_events.c: remove unneeded inclusion of list.h

        * ircd/whocmds.c (do_who): hide server name in /who unless
        requester is an operator; simplify hop count insertion

        * ircd/s_misc.c (exit_one_client): make sure client's snomask is
        cleared

        * ircd/parse.c: use mo_version and mo_admin when opers do /version
        or /admin

        * ircd/m_whowas.c (m_whowas): use HEAD_IN_SAND_SERVERNAME instead
        of the static string "*.undernet.org"

        * ircd/m_version.c: only let ordinary users get version
        information for the server they are on

        * ircd/m_admin.c: only let ordinary users get admin information
        for the server they are on

        * ircd/channel.c (client_can_send_to_channel): check is_banned()
        before letting the client speak on a channel s/he is not on

        * include/supported.h: add NETWORK to feature list

        * include/handlers.h: declare mo_admin() and mo_version()

2002-01-10  Kevin L Mitchell  <klmitch@mit.edu>

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/s_debug.c (count_memory): conditionalize on MDEBUG instead
        of !NDEBUG

        * ircd/m_stats.c: conditionalize /stats M on MDEBUG instead of
        !NDEBUG

        * ircd/ircd_alloc.c: conditionalize on MDEBUG instead of on
        !NDEBUG

        * ircd/fda.c: conditionalize on MDEBUG instead of on !NDEBUG

        * ircd/Makefile.in: run make depend on chkconf.c as well

        * include/ircd_alloc.h: instead of conditionalizing on !NDEBUG,
        conditionalize on MDEBUG

        * include/fda.h: instead of conditionalizing on !NDEBUG,
        conditionalize on MDEBUG

        * configure: rebuild configure script

        * configure.in: enable assertion checking by default, since we
        have now decoupled memory debugging from the NDEBUG macro

        * ircd/s_user.c (set_nick_name): remove calls to
        verify_client_list()

        * ircd/s_misc.c (exit_one_client): remove calls to
        verify_client_list()

        * ircd/s_conf.c (rehash): remove calls to verify_client_list()

        * ircd/m_who.c (m_who): remove calls to verify_client_list()

        * ircd/list.c: remove calls to verify_client_list(); keep
        verify_client_list() around just in case we ever need it again,
        but never compile it in

        * ircd/ircd_events.c (event_execute): remove calls to
        verify_client_list()

        * ircd/client.c (client_get_ping): remove calls to
        verify_client_list()

        * include/list.h (send_listinfo): remove temporary debugging
        function verify_client_list()

        * ircd/uping.c: don't die if the event type is ET_ERROR in socket
        callback functions

        * ircd/res.c (res_callback): don't die if the event type is
        ET_ERROR

        * ircd/listener.c (accept_connection): don't die if the event type
        is ET_ERROR

2002-01-09  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/s_user.c (set_nick_name): bracket call to
        add_client_to_list() with calls to verify_client_list()

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel (again)

        * ircd/list.c (verify_client_list): add a probabilistic loop
        detector: for every client added, there is a 2% probability that
        it will be used to replace the value of sentinel; if at any time,
        sentinel is found again, we know we're in a loop

        * ircd/ircd_events.c (event_execute): add verify_client_list()
        calls wrapping event_execute; at the very least, I'll figure out
        what event the corruption occurred in

        * ircd/list.c: moved verify_client_list() to try to keep it from
        being inlined

        * ircd/Makefile.in (version.c): version.c wasn't dependant on
        version.h and patchlevel.h, like it was supposed to be

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/s_auth.c (destroy_auth_request): overload send_reports
        argument to also indicate whether or not to call
        release_auth_client() and thereby enter the client into the linked
        list

        * ircd/engine_devpoll.c (engine_loop): remove bogus assertion

        * include/patchlevel.h (PATCHLEVEL): bump patchlevel

        * ircd/list.c (free_client): verify that destroy_auth_request()
        didn't automagically re-add us to the list; we might have to think
        about this interaction more carefully, actually

        * ircd/s_auth.c (auth_kill_client): zero the auth pointer before
        calling free_client(); otherwise, free_client() will try to free
        the auth by calling destroy_auth_request(), which will call
        add_client_to_list()

        * ircd/s_misc.c (exit_one_client): liberally sprinkle calls to
        verify_client_list() around to catch any corruption that might
        occur here

        * ircd/s_conf.c (rehash): liberally sprinkle calls to
        verify_client_list() here, since this is about the only routine I
        can think of that could cause the "core on kill -HUP" bug

        * ircd/m_who.c: sprinkle calls to verify_client_list() around
        liberally, since we've seen crashes here; temporarily include the
        otherwise unneeded list.h header

        * ircd/list.c: sprinkle calls to verify_client_list() around quite
        liberally; add debugging asserts to list manipulation functions to
        catch strange settings for next and prev pointers; define
        verify_client_list(), which walks the client list and verifies
        that everything is as it's supposed to be

        * ircd/client.c: wrap client_get_ping with calls to
        verify_client_list() to see if that's where we're dying

        * include/patchlevel.h (PATCHLEVEL): bump to 03

        * include/list.h: declare verify_client_list() if DEBUGMODE
        enabled; otherwise, define it to be empty

2002-01-08  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/m_quit.c (m_quit): remove an unused variable

        * include/patchlevel.h (PATCHLEVEL): bump PATCHLEVEL to 2

        * ircd/s_user.c: when building the user mode to send to the user,
        don't include +r; add an extra set of parens to squelch a warning

        * ircd/m_quit.c (m_quit): use exit_client_msg()

        * include/patchlevel.h (PATCHLEVEL): bump patch level, so we can
        keep track of who's running what version

        * ircd/m_squit.c (ms_squit): remove debugging calls to
        protocol_violation()

        * Makefile.in: change MAKEFILES to IRCD_MAKEFILES to work around a
        new gmake "feature" (pull-up from trunk)

        * ircd/m_quit.c (m_quit): prefix user quits with "Quit:" (pull-up
        from trunk)

2002-01-07  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/s_user.c: add FLAGS_ACCOUNT, represented as 'r', to the
        list of user modes; process account name as part of user mode in
        NICK decoding (set_nick_name()); add account name to usermode when
        building the usermode to incorporate in outgoing NICK messages

        * ircd/s_err.c: add RPL_WHOISACCOUNT for reporting what account a
        user is logged in to

        * ircd/parse.c: define the new ACCOUNT command, usable only by
        servers and ignored by everything else

        * ircd/m_whois.c: report what account name is associated with a
        user, if any

        * ircd/m_account.c: implement the ACCOUNT command

        * ircd/Makefile.in: add m_account.c to the list of sources; ran
        make depend

        * include/struct.h: add an account field to struct User

        * include/numeric.h: add a reply, RPL_WHOISACCOUNT, for reporting
        what username a user is logged in under

        * include/msg.h: add ACCOUNT command and token (AC)

        * include/ircd_defs.h: define ACCOUNTLEN to be 12--this matches
        the maximum length of a username for X

        * include/handlers.h: add declaration for ms_account()

        * include/client.h: add FLAGS_ACCOUNT to flag when a user account
        name has been set; added FLAGS_ACCOUNT to SEND_UMODES; added
        IsAccount() and SetAccount() to manipulate the flag

        * ircd/m_squit.c (ms_squit): if we call FindNServer() on a server
        name like "Amsterdam2.NL.EU.undernet.org", we get the struct
        Client for the server with numeric "Am", which happens to be
        stockholm!  To fix this, we look up the full name *first*; if that
        doesn't get it, *then* we look up by numeric.

2001-12-24  Perry Lorier <isomer@coders.net>
        * ircd/m_server.c: cleanups, maybe this will make the bug easier
        to find.

        * ircd/m_stats.c: display maximum number of connects in an I:

2001-11-22  Perry Lorier  <isomer@coders.net>
        * ircd/m_squit.c: Bug fix in squit

2001-11-03  Greg Sikorski <gte@atomicrevs.demon.co.uk>
        * ircd/parse.c, include/handlers.h: Give remote whois the correct
        handler.
        
2001-11-01  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/send.c: some minor white-space fiddling; recast selector
        test in sendwallto_group_butone() to remove a warning regarding
        putting & within parentheses

        * ircd/m_create.c (ms_create): use time_t instead of int as a
        declaration for rate

        * ircd/ircd_reply.c (protocol_violation): it's supposed to be
        WALL_DESYNCH, not CMD_DESYNCH, if I understand things right--no
        wonder we weren't seeing any protocol violations!

        * include/send.h: include time.h for time_t; move WALL_* closer to
        the function they're used in; some white-space fiddling; add
        declaration of sendto_opmask_butone_ratelimited()

        * ircd/m_squit.c (ms_squit): add protocol_violation() calls in the
        cases where we ignore a squit, so we aren't taken by surprise, at
        least...

        * ircd/m_create.c (ms_create): Display origin server, not origin
        user

        * ircd/m_create.c (ms_create): Fix "Timestamp drift" server notice

2001-10-31  Perry Lorier  <isomer@coders.net>
        * include/m_ping.c: Forward port ping bug

2001-10-31  Perry Lorier  <isomer@coders.net>
        * include/patchlevel.h: We're beta now

2001-10-31  Perry Lorier  <isomer@coders.net>
        * ircd/s_user.c: fixed hunt_server

2001-09-21  Perry Lorier  <isomer@coders.net>
        * ircd/send.c and various: replace sendcmdto_flag_butone with
        sendwallto_group_butone

2001-09-21  Vampire-  <unknown>
        * ircd/ircd_string.c: unique_name_vector round II.

2001-09-21  mbuna  <mbuna@undernet.org>
        * configure.in: Add support for darwin

2001-09-21  Perry Lorier  <isomer@coders.net>
        * ircd/s_user.c I'm stupid, s/acptr/from/, Hektik pointed it out

2001-09-20  Perry Lorier  <isomer@coders.net>

        * Pullups from 2.10.10.pl16
        * Added some warnings, and the concept of rate limited snotices

2001-08-31  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/channel.c: use "%u" to format limit arguments; use
        strtoul() to process limit arguments in a /mode command--note:
        most clients seem to truncate the integer, probably because
        they're using atoi, and perhaps signed ints

2001-08-17  Kevin L Mitchell  <klmitch@mit.edu>

        * ircd/numnicks.c: include stdlib.h for exit()

        * ircd/ircd_log.c: include stdlib.h for exit()

        * ircd/ircd_events.c: include stdlib.h for exit()

        * ircd/s_stats.c: remove description of /stats v, since it's gone

        * ircd/m_wallops.c (mo_wallops): add "*" to the beginning of
        /wallops to distinguish wallops from wallusers

        * ircd/m_error.c (mr_error): ignore ERROR from clients that aren't
        in the "handshake" or "connecting" states--I think the latter will
        never happen, but...

        * doc/Authors: apply delete's Authors patch

        * RELEASE.NOTES: rewrite RELEASE.NOTES, basing it a little on
        Braden's version

        * README: rewrite README

2001-07-31  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_serv.c (server_estab): remove unused variable split

        * ircd/parse.c: add mr_error to the parse table

        * ircd/m_error.c (mr_error): add mr_error() to handle ERRORs from
        unregistered connections--if IsUserPort() is true, the ERROR is
        ignored, otherwise, the message is saved

2001-07-28  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_kill.c (ms_kill): another minor typo *sigh*

        * ircd/s_user.c (send_supported): oops, minor typo...

        * ircd/s_user.c: implement send_supported() to send two ISUPPORT
        messages containing our feature buffers; make register_user() use
        send_supported()

        * ircd/s_misc.c (exit_client): make sure not to give away a remote
        server in the ERROR message sent to the client; if the killer is a
        server, we substitute our name in its place

        * ircd/m_version.c (m_version): use send_supported() to send the
        ISUPPORT values to the user

        * ircd/m_nick.c: shave nick collision kills here a bit, too, for
        the same reasons as for m_kill.c

        * ircd/m_kill.c: shave kills a bit so that the results look
        exactly the same no matter where you are; if we didn't do this, it
        would be possible to map the network by looking at the differences
        between kills originating under various circumstances

        * include/supported.h: split the features into two, so as to not
        bust the parameter count when sending the features list

        * include/s_user.h: declare new send_supported() function to send
        the ISUPPORT information

2001-07-27  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_bsd.c: disable IP (*not* TCP) options to prevent
        source-routed spoofing attacks; this is only available under
        u2.10.11, so don't even bother, since no one but testers are using
        the source base

2001-07-25  Kevin L. Mitchell  <klmitch@mit.edu>

        * include/ircd_policy.h: enable HEAD_IN_SAND_REMOTE by default

        * ircd/s_err.c: put in a . for reporting link version on /trace,
        to match what /version does

2001-07-21  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_misc.c (exit_client): servers don't understand what the
        numeric nick ERROR is supposed to mean, so they ignore error
        messages, resulting in not knowing why we were rejected; use
        sendcmdto_one for servers and sendrawto_one for clients

2001-07-17  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_burst.c (ms_burst): in the case of a modeless channel and
        a nick collide, a bare BURST may be propagated; adjust the
        enforced parameter count to accept the bare BURST

2001-07-12  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_bsd.c: mark a client as having been IP checked

        * ircd/IPcheck.c (ip_registry_check_remote): remove unneeded
        second call to SetIPChecked()

2001-07-11  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/engine_poll.c: deal with POLLHUP properly (hopefully)

        * ircd/engine_devpoll.c: deal with POLLHUP properly (hopefully)

2001-07-09  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/os_bsd.c (os_get_rusage): move buf into the two ifdef'd
        sections so that if neither is used, the declaration of buf will
        not elicit an "unused variable" warning under NetBSD

        * ircd/m_map.c: include string.h to declare strcpy (fix warnings
        on alpha)

        * ircd/m_away.c: include string.h to declare strcpy/strlen (fix
        warnings on alpha)

        * ircd/ircd_log.c: include string.h to declare strcpy/strlen (fix
        warnings on alpha)

        * ircd/client.c: include string.h to declare memset (fix warnings
        on alpha)

        * ircd/channel.c: remove unused functions next_overlapped_ban,
        del_banid, and is_deopped (fix warnings under -O1)

        * ircd/IPcheck.c: include string.h to declare memset/memcpy (fix
        warnings on alpha)

2001-06-29  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_user.c (set_user_mode): clear the snomask if the user
        isn't supposed to receive server notices anymore

        * ircd/ircd_features.c: change CONFIG_OPERCMDS to default to FALSE

        * configure.in: use AC_MSG_CHECKING/AC_MSG_RESULT when checking
        installation prefix; default devpoll and kqueue to on (they get
        turned off if the required headers aren't present)

        * ircd/whocmds.c (do_who): use ircd_snprintf() instead of
        sprintf_irc(); it's a bit hackish, but it'll do for now

        * ircd/support.c: remove unused #include

        * ircd/send.c: remove unused #include

        * ircd/s_user.c: use ircd_snprintf() instead of sprintf_irc()

        * ircd/s_serv.c: remove unused #include

        * ircd/s_misc.c: use ircd_snprintf() and friends instead of
        sprintf_irc() and friends

        * ircd/s_err.c: moved atoi_tab[] from ircd/sprintf_irc.c to
        ircd/s_err.c, which is the only other file to refer to it

        * ircd/s_conf.c (conf_add_deny): use ircd_snprintf() instead of
        sprintf_irc()

        * ircd/s_bsd.c (connect_server): use ircd_snprintf() instead of
        sprintf_irc()

        * ircd/s_auth.c: use ircd_snprintf() instead of sprintf_irc()

        * ircd/res.c: use ircd_snprintf() instead of sprintf_irc()

        * ircd/m_version.c: use ircd_snprintf() instead of sprintf_irc()

        * ircd/m_kill.c: use ircd_snprintf() instead of sprintf_irc()

        * ircd/listener.c: use ircd_snprintf() instead of sprintf_irc()

        * ircd/gline.c: use ircd_snprintf() instead of sprintf_irc()

        * ircd/channel.c: don't include sprintf_irc.h; use ircd_snprintf()
        instead of sprintf_irc()

        * ircd/Makefile.in: remove sprintf_irc.c from sources list; run
        make depend

        * include/ircd_string.h: remove declaration of sprintf_irc() (what
        was it doing here anyway?)

        * include/sprintf_irc.h: removed unneeded source file

        * ircd/sprintf_irc.c: removed unneeded source file

        * ircd/s_debug.c (count_memory): remove some dead code

        * ircd/s_auth.c: remove some dead code

        * ircd/res.c (update_list): remove some dead code

        * ircd/m_whowas.c: remove some dead code

        * ircd/m_whois.c: remove some dead code

        * ircd/m_who.c: remove some dead code

        * ircd/m_wallusers.c: remove some dead code

        * ircd/m_wallops.c: remove some dead code

        * ircd/m_wallchops.c: remove some dead code

        * ircd/m_version.c: remove some dead code

        * ircd/m_userip.c: remove some dead code

        * ircd/m_userhost.c: remove some dead code

        * ircd/m_uping.c: remove some dead code

        * ircd/m_trace.c: remove some dead code

        * ircd/m_topic.c: remove some dead code

        * ircd/m_tmpl.c: remove some dead code

        * ircd/m_time.c: remove some dead code

        * ircd/m_squit.c: remove some dead code

        * ircd/m_silence.c: remove some dead code

        * ircd/m_settime.c: remove some dead code

        * ircd/m_set.c: remove some dead code

        * ircd/m_server.c: remove some dead code

        * ircd/m_rpong.c: remove some dead code

        * ircd/m_rping.c: remove some dead code

        * ircd/m_restart.c: remove some dead code

        * ircd/m_reset.c: remove some dead code

        * ircd/m_rehash.c: remove some dead code

        * ircd/m_quit.c: remove some dead code

        * ircd/m_proto.c: remove some dead code

        * ircd/m_privs.c: remove some dead code

        * ircd/m_privmsg.c: remove some dead code

        * ircd/m_pong.c: remove some dead code

        * ircd/m_ping.c: remove some dead code

        * ircd/m_pass.c: remove some dead code

        * ircd/m_part.c: remove some dead code

        * ircd/m_opmode.c: remove some dead code

        * ircd/m_oper.c: remove some dead code

        * ircd/m_notice.c: remove some dead code

        * ircd/m_nick.c: remove some dead code

        * ircd/m_map.c: remove some dead code

        * ircd/m_lusers.c: remove some dead code

        * ircd/m_list.c: remove some dead code

        * ircd/m_links.c: remove some dead code

        * ircd/m_kill.c: remove some dead code

        * ircd/m_kick.c: remove some dead code

        * ircd/m_jupe.c: remove some dead code

        * ircd/m_join.c: remove some dead code

        * ircd/m_ison.c: remove some dead code

        * ircd/m_invite.c: remove some dead code

        * ircd/m_info.c: remove some dead code

        * ircd/m_help.c: remove some dead code

        * ircd/m_gline.c: remove some dead code

        * ircd/m_get.c: remove some dead code

        * ircd/m_error.c: remove some dead code

        * ircd/m_endburst.c: remove some dead code

        * ircd/m_die.c: remove some dead code

        * ircd/m_desynch.c: remove some dead code

        * ircd/m_destruct.c: remove some dead code

        * ircd/m_defaults.c: remove some dead code

        * ircd/m_create.c: remove some dead code, along with an #if 1

        * ircd/m_cprivmsg.c: remove some dead code

        * ircd/m_connect.c: remove some dead code

        * ircd/m_close.c: remove some dead code

        * ircd/m_clearmode.c: remove some dead code

        * ircd/m_burst.c: remove some dead code

        * ircd/m_away.c: remove some dead code

        * ircd/m_admin.c: remove some dead code

        * ircd/listener.c (accept_connection): remove some dead code

        * ircd/ircd_reply.c (need_more_params): remove some dead code

        * ircd/channel.c (add_banid): remove some dead code

        * include/support.h: remove some dead code

        * include/querycmds.h: remove some dead code

        * doc/readme.chroot: document how to do chroot operation

2001-06-28  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/Makefile.in: tune for VPATH builds/installs; add a rule to
        force bin directory to be created if necessary prior to
        installation; run make depend

        * doc/Makefile.in (install): tune for VPATH installs by cd'ing to
        the ${srcdir}

        * Makefile.in: tune to detect Makefile.in changes in
        subdirectories and to create installation directory indicated by
        ${prefix}

        * ircd/whocmds.c (count_users): routine to count the number of
        users matching a given user@host mask

        * ircd/s_err.c: add error messages for ERR_LONGMASK,
        ERR_TOOMANYUSERS, and ERR_MASKTOOWIDE

        * ircd/m_gline.c: look for and advance past '!' flag on G-lines
        from operators; only set GLINE_OPERFORCE flag if oper has the
        PRIV_WIDE_GLINE privilege

        * ircd/ircd_features.c: add GLINEMAXUSERCOUNT, which is the
        maximum number of users a G-line can impact before it has to be
        forced; OPER_WIDE_GLINE, to allow operators to use ! to force a
        wide G-line to be set; and LOCOP_WIDE_GLINE, to allow local
        operators to use ! to force a wide G-line to be set

        * ircd/gline.c: make make_gline() be called with separate user and
        host arguments, and not call canon_userhost() directly; implement
        gline_checkmask() to verify that a host mask is acceptable; move
        BADCHAN check up in gline_add(), and check passed-in mask under
        certain circumstances for acceptability; fix call to
        sendto_opmask_butone() to handle separation of userhost into user
        and host in gline_add(); update call to make_gline()

        * ircd/client.c: use FEAT_OPER_WIDE_GLINE and
        FEAT_LOCOP_WIDE_GLINE to set PRIV_WIDE_GLINE for an operator; add
        PRIV_WIDE_GLINE to privtab[] for client_report_privs()

        * include/whocmds.h (count_users): declare routine to count users
        matching a given user@host mask

        * include/numeric.h: added three new error returns: ERR_LONGMASK
        -- mask can't be formatted into a buffer; ERR_TOOMANYUSERS -- too
        many users would be impacted by the mask; ERR_MASKTOOWIDE -- mask
        contains wildcards in the wrong places

        * include/ircd_features.h: add FEAT_GLINEMAXUSERCOUNT,
        FEAT_OPER_WIDE_GLINE, and FEAT_LOCOP_WIDE_GLINE

        * include/gline.h (GLINE_OPERFORCE): provides a way for m_gline()
        to signal to gline_add() that the operator attempted to force the
        G-line to be set

        * include/client.h (PRIV_WIDE_GLINE): new privilege for operators

        * doc/readme.gline: update to document new "!" prefix to a G-line
        user@host mask

        * doc/readme.features: document GLINEMAXUSERCOUNT,
        OPER_WIDE_GLINE, and LOCOP_WIDE_GLINE

        * doc/example.conf: update to mention new features along with
        their defaults

2001-06-27  Kevin L. Mitchell  <klmitch@mit.edu>

        * doc/example.conf: updated example.conf from Braden
        <dbtem@yahoo.com>

        * include/supported.h: forward-port from pl15

2001-06-25  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/whocmds.c: include ircd_policy.h and implement
        HEAD_IN_SAND_WHO_OPCOUNT--forward-port from pl15

        * ircd/m_whois.c: forward-port of the idle-time hiding code from
        pl15; this also required passing parc into do_whois(), which also
        meant passing parc into do_wilds()--*sigh*

        * include/ircd_policy.h: add a couple more HEAD_IN_SAND
        #define's--WHOIS_IDLETIME and WHO_HOPCOUNT

2001-06-22  Kevin L. Mitchell  <klmitch@mit.edu>

        * tools/wrapper.c: add a wrapper program that can be used to
        adjust file descriptor limits and root directories; program must
        be run as root--NOT SETUID!--and given appropriate -u arguments

        * doc/readme.log: documentation of how to configure logging

        * doc/readme.features: documentation of each feature (except for
        logging)

2001-06-21  Kevin L. Mitchell  <klmitch@mit.edu>

        * Makefile.in (config): add a deprecation notice with a pointer to
        tools/transition

        * tools/transition: shell script to convert old compile-time
        options into new compile-time options and appropriate F-lines

        * tools/mkchroot: shell-script to prepare the chroot area by
        copying over all the necessary libraries so they can be found

2001-06-20  Kevin L. Mitchell  <klmitch@mit.edu>

        * INSTALL: partial update of INSTALL for u2.10.11 release...

2001-06-14  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/table_gen.c (makeTables): finally got tired of the
        "overflow in implicit conversion" warning, so just got rid of it
        by explicitly casting UCHAR_MAX to a (default) char; diffs show no
        differences in the tables generated

2001-06-11  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/send.c (sendcmdto_match_butone): don't let the server crash
        if a client is in the STAT_CONNECTING status

2001-06-10  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/send.c: remove unused vsendcmdto_one(), consolidating it
        into sendcmdto_one(); define new sendcmdto_prio_one(), which
        places the message into the priority queue

        * ircd/s_user.c (hunt_server_prio_cmd): definition of
        hunt_server_prio_cmd(), which simply calls sendcmdto_prio_one()
        instead of sendcmdto_one()

        * ircd/m_settime.c: use sendcmdto_prio_one() and
        hunt_server_prio_cmd() to send SETTIME

        * ircd/m_server.c: use sendcmdto_prio_one() to send SETTIME

        * include/send.h: removed declaration for unused vsendcmdto_one();
        added a declaration for sendcmdto_prio_one()

        * include/s_user.h: declare hunt_server_prio_cmd(), which calls
        sendcmdto_prio_one()

        * ircd/send.c (sendcmdto_flag_butone): oops; /wallops should be
        put in the server's priority queue, too...

        * ircd/ircd.c: don't check LPATH for accessibility at all

2001-06-08  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_serv.c (server_estab): send a +h flag in our SERVER
        command if we're configured as a hub; send individual server flags
        in SERVER commands

        * ircd/s_bsd.c (completed_connection): send a +h flag in our
        SERVER command if we're configured as a hub

        * ircd/m_server.c: implement parv[7] as a mode-like string; +h
        sets the FLAGS_HUB flag for a server; +s sets the FLAGS_SERVICE
        flag for a server; +hs sets both flags; also modify CMD_SERVER
        format string to send the flags

        * include/client.h: define two new flags, FLAGS_HUB and
        FLAGS_SERVICE to mark services and hubs as such; define testing
        macros, setting macros

        * ircd/s_user.c: remove deprecated struct Gline* argument to
        register_user(); remove GLINE rebroadcast; do not send GLINE
        acknowledgement parameter to NICK; do not look for GLINE
        acknowledgement parameter to NICK while parsing

        * ircd/s_serv.c (server_estab): remove deprecated struct Jupe*
        argument to server_estab(); do not send JUPE/GLINE acknowledgement
        parameters for SERVER or NICK

        * ircd/m_user.c (m_user): remove deprecated argument to
        register_user()

        * ircd/m_server.c: remove deprecated argument to server_estab();
        remove documentation comment regarding JUPE acknowledgement
        parameter to SERVER; remove JUPE rebroadcast

        * ircd/m_pong.c (mr_pong): remove deprecated argument to
        register_user()

        * ircd/m_nick.c: remove documentation comment regarding GLINE
        acknowledgement parameter to NICK

        * ircd/jupe.c: use user's real name in JUPE server notices if
        HEAD_IN_SAND_SNOTICES is defined

        * ircd/ircd.c: remove deprecated chroot() code; remove deprecated
        setuid code; correct ancient DEBUG vs DEBUGMODE typo

        * ircd/gline.c: use user's real name in GLINE server notices if
        HEAD_IN_SAND_SNOTICES is defined

        * ircd/channel.c (modebuf_flush_int): make apparent source be
        local server, not oper's server; use user's real name in hack
        notices and DESYNC notices if HEAD_IN_SAND_SNOTICES is defined

        * include/s_user.h: remove struct Gline pre-declaration; remove
        deprecated struct Gline argument from register_user()

        * include/s_serv.h: remove struct Jupe pre-declaration; remove
        deprecated struct Jupe argument from server_estab()

2001-06-07  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_stats.c (hunt_stats): forward-port from pl15 of all the
        changes required to control remote stats

        * ircd/s_numeric.c (do_numeric): rewrite numeric origins if
        recipient is not an operator and HEAD_IN_SAND_REWRITE is defined
        [forward-port from pl15]

        * ircd/m_whowas.c (m_whowas): report server name only if requester
        is an operator [forward-port from pl15]

        * ircd/m_whois.c (do_whois): /whois <mynick> now correctly reports
        my server; if HEAD_IN_SAND_REMOTE is 1, ignore the middle argument
        and obtain the report from the user's server [forward-port from
        pl15]

        * ircd/m_who.c: add missing include for ircd_policy.h
        [forward-port from pl15]

        * ircd/m_version.c (m_version): require oper access for remote
        /version if HEAD_IN_SAND_REMOTE is 1 [forward-port from pl15]

        * ircd/m_time.c (m_time): require oper access for remote /time if
        HEAD_IN_SAND_REMOTE is 1 [forward-port from pl15]

        * ircd/m_stats.c: pass extra argument to hunt_stats(); correct
        missing semicolon [forward-port from pl15]

        * ircd/m_nick.c (ms_nick): hide the origin of certain collision
        kills [forward-port from pl15]

        * ircd/m_motd.c (m_motd): require oper access for remote /motd if
        HEAD_IN_SAND_REMOTE is 1 [forward-port from pl15]

        * ircd/m_lusers.c (m_lusers): require oper access for remote
        /lusers if HEAD_IN_SAND_REMOTE is 1 [forward-port from pl15]

        * ircd/m_burst.c (ms_burst): server-added bans are stored using
        local server name, to hide remote server names; modes also are to
        originate from the local server [forward-port from pl15]

        * ircd/m_admin.c (m_admin): require oper access for remote /admin
        if HEAD_IN_SAND_REMOTE is 1 [forward-port from pl15]

        * ircd/channel.c (add_banid): if a server is adding a ban, use my
        server name to hide the remote server's name [forward-port from
        pl15]

        * ircd/Makefile.in: ran make depend

        * include/s_stats.h: hunt_stats() has to have an extra argument to
        support the forward-port from pl15

        * include/ircd_policy.h: #define HEAD_IN_SAND_STATS_P; add
        HEAD_IN_SAND_{BANWHO,REWRITE,REMOTE} [forward-port from pl15]

        * ircd/engine_poll.c (engine_loop): remove bogus assert that I
        forgot to check in the events branch

2001-06-06  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/res.c (init_resolver): don't start DNS expires with a 0
        relative timeout--if the server starts slow, timeouts could be
        messy...there's probably a better solution, but this'll do for now

        * ircd/os_solaris.c: _XOPEN_SOURCE doesn't get along with Solaris
        headers very well; include stropts.h; define an os_set_tos()

        * ircd/os_generic.c (os_set_tos): added an os_set_tos() for
        os_generic.c

        * ircd/ircd.c: if there are no C-lines, we don't want to have a
        timer that expires at the absolute time of 0--it kinda blocks all
        the other timers!

        * ircd/engine_devpoll.c: some includes for open(); declare errcode
        and codesize in engine_loop()

        * ircd/list.c (free_client): remove bogus check on timer active
        flag

        * ircd/s_auth.c: pull out destruction code in
        auth_timeout_request() into an externally-visible
        destroy_auth_request(); manage cli_auth pointer in client
        structure; use it for an extra assertion check

        * ircd/list.c: include s_auth.h for destroy_auth_request(); add
        debugging notices to show flow when deallocating
        connections/clients; call destroy_auth_request() when free'ing a
        client that has an auth outstanding; don't free the connection if
        the process timer is unmarked but still active

        * ircd/ircd_events.c: set GEN_ACTIVE when initializing a generator
        and reset it before calling the event handler for an ET_DESTROY
        event

        * include/s_auth.h (destroy_auth_request): declare
        destroy_auth_request(), which can be used to destroy an
        outstanding auth request if a client socket goes away before the
        auth exchange is completed

        * include/ircd_events.h: add an active flag to keep track of
        whether or not particular generators are active, for the
        convenience of functions using the API

        * include/client.h: add a pointer for auth requests to struct
        Connection so we can kill outstanding auth requests if a client
        socket closes unexpectedly

        * ircd/s_bsd.c: cli_connect() could become 0 during the course of
        the sock or timer callback; take that into account in the assert

        * ircd/list.c: add magic number checking and setting--magic
        numbers are zero'd on frees to detect double-frees; add back
        setting of cli_from() to 0 to break the back-link from the struct
        Connection (duh)

        * ircd/ircd.c: set me's magic number correctly

        * include/client.h: define magic numbers and accessor/verifier
        macros

        * ircd/list.c: assert that dealloc_client() is called with
        cli_connect(cptr) == 0; set cli_connect(cptr) to 0 before calling
        dealloc_client(); don't mess with cli_from(cptr)

        * ircd/s_bsd.c: only attempt to dealloc a connection if the
        associated client has already been destroyed, or at least delinked

2001-06-05  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/list.c (free_client): only try to delete the socket when
        the fd hasn't already been closed, avoiding a double-free

        * ircd/list.c (free_connection): make sure the client is really
        gone before doing away with the connection

        * ircd/s_bsd.c: record that socket has been added in con_freeflag
        field; queue a socket_del() as soon as the socket is close()'d;
        use con_freeflag & FREEFLAG_TIMER instead of con_timer; clear
        FREEFLAG_SOCKET on ET_DESTROY event in client_sock_callback(),
        then dealloc the connection if safe; mark socket as dead when
        there's a read error or EOF; clear FREEFLAG_TIMER flag upon entry
        to client_timer_callback(); dealloc connection if safe upon
        ET_DESTROY event in client_timer_callback()

        * ircd/list.c: use con_freeflag instead of con_timer; only dealloc
        the connection if both socket and timer have been destroyed;
        destroy both socket and timer explicitly and carefully

        * include/client.h: replace the con_timer field with a
        con_freeflag field, to indicate what still needs freeing; define
        the freeflags

        * ircd/engine_select.c (engine_loop): duh...sockList[i] could
        become 0

        * ircd/engine_devpoll.c (engine_loop): duh...sockList[i] could
        become 0

        * ircd/s_bsd.c: add some extra assertions to try to track down a
        corruption problem

        * ircd/engine_select.c (engine_loop): add an extra assert to try
        to track down a corruption problem

        * ircd/engine_poll.c (engine_loop): add an extra assert to try to
        track down a corruption problem

        * ircd/engine_kqueue.c (engine_loop): add an extra assert to try
        to track down a corruption problem

        * ircd/engine_devpoll.c (engine_loop): skip slots that have become
        empty during processing; add an extra assert to try to track down
        a corruption problem

        * ircd/engine_kqueue.c (engine_delete): make sure to zero deleted
        entries

2001-06-04  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_bsd.c (client_sock_callback): client is no longer
        blocked, so we must mark it as unblocked

        * ircd/engine_select.c: add Debug() calls galore; add handling for
        SS_NOTSOCK; use a dummy sock variable to keep things from
        disappearing on us; correct timeout calculation; update nfds for
        efficiency

        * ircd/engine_poll.c: use new debugging level (DEBUG_ENGINE);
        remove a spurious "if (sock)" which will always be true; update
        nfds for efficiency

        * ircd/engine_kqueue.c: add Debug() calls galore; add handling for
        SS_NOTSOCK (just in case); correct timeout calculation

        * ircd/engine_devpoll.c: add Debug() calls galore; add handling
        for SS_NOTSOCK; correct timeout calculation; add EAGAIN handling

        * include/s_debug.h (DEBUG_ENGINE): add new debugging level;
        pretty-indent numbers

        * ircd/engine_poll.c (engine_loop): break out SS_NOTSOCK
        case--it's not a socket; the check for writability is most likely
        not needed, but present for completeness

2001-05-24  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_bsd.c: add Debug messages; call read_packet() even if the
        no newline flag is set; call read_packet() when the timer expires,
        regardless of what's in the buffer--read_packet() should be able
        to deal properly

        * ircd/IPcheck.c (ip_registry_connect_succeeded): correct a NOTICE
        sent to clients to include the client nickname (duh)

        * ircd/ircd_events.c: don't destroy a timer if it's already marked
        for destruction; replace a missing ! in socket_del()

        * ircd/engine_poll.c (engine_loop): reference a temporary variable
        so we don't have to worry about sockList[i] going away

        * ircd/s_bsd.c (client_sock_callback): add Debug messages

        * ircd/s_auth.c: add Debug messages all over the place

        * ircd/ircd_events.c: add and edit some Debug messages; add a list
        of routines to convert some of the enums and flags from numbers
        into human-readable strings for the Debug messages

        * ircd/engine_poll.c: hack some Debug messages to use the new name
        conversion routines in ircd_events.c; add an extra assert for a
        condition that shouldn't ever happen; apparently recv() can return
        EAGAIN when poll() returns readable--I wonder why...

        * include/ircd_events.h: declare some helper routines under
        DEBUGMODE

2001-05-23  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_bsd.c (client_sock_callback): add an extra assertion
        check

        * ircd/s_auth.c: add more Debug messages

        * ircd/list.c (make_client): add an extra assertion check

        * ircd/ircd_events.c (socket_events): don't call the engine events
        changer if we haven't actually made any changes to the event mask

        * ircd/uping.c: add some Debug messages

        * ircd/s_stats.c: document new /STATS e

        * ircd/s_err.c: add RPL_STATSENGINE to report the engine name

        * ircd/s_bsd.c: remove static client_timer variable; in
        read_packet(), if there's still data in the client's recvQ after
        parsing, add a 2 second timer (con_proc); fix the ET_DESTROY case
        of client_sock_callback to handle destroying the timer properly;
        rewrote client_timer_callback from scratch to be called on an
        individual client

        * ircd/m_stats.c: add /STATS e to report the engine name

        * ircd/list.c: deal with con_timer field in struct Connection
        properly; correct a core-level bug in remove_client_from_list--if
        the client is the only one in the list, we try to update
        GlobalClientList's cli_prev pointer--not good

        * ircd/ircd.c: remove call to init_client_timer()

        * ircd/engine_poll.c: made Debug messages more uniform by
        prepending "poll:" to them all; corrected an off-by-one error that
        caused poll_count to be 1 less than the actual count and removed
        my work-around; added Debug messages to indicate which socket is
        being checked and what the results are

        * ircd/Makefile.in: ran a make depend

        * include/s_bsd.h: remove init_client_timer(), since we're doing
        it differently now

        * include/numeric.h (RPL_STATSENGINE): a stats reply to report the
        engine name

        * include/ircd_policy.h (HEAD_IN_SAND_STATS_E): turn off /stats e
        reports for non-opers

        * include/client.h: add con_timer and con_proc fields to struct
        Connection and define accessor macros--con_timer marks that
        con_proc contains a valid timer, and con_proc is used to pace user
        data

        * ircd/s_bsd.c (close_connection): let free_client() destroy the
        socket

        * ircd/s_auth.c (start_auth): add a Debug call to indicate when
        auth has begun on a client

        * ircd/ircd_events.c: ensure that event_execute() is called with a
        non-NULL event; modify event_add() macro to properly zero list
        bits; modify gen_dequeue() to not try to clip it out of a list
        it's already been clipped out of; change signal socket
        initialization to use state SS_NOTSOCK; permit timeout values of
        0 in add_timer(); add many Debug calls; change socket_del() and
        timer_del() to always set the GEN_DESTROY flag; use GEN_MARKED in
        timer_run() instead of GEN_DESTROY so that event_generate() will
        pass on the events; remove the switch and replace with a simpler
        if-then-else tree in timer_run(); don't allow destroyed sockets to
        be destroyed again, nor their states or event masks to be changed

        * ircd/ircd.c: initialize "running" to 1

        * ircd/engine_poll.c: deal with SS_NOTSOCK "sockets"; add Debug
        messages all over the place; fix a counting problem in
        engine_add(); turn wait into a signed integer and set it to -1
        only if timer_next() returns 0; adjust wait time to be relative;
        don't call gen_ref_dec() if socket disappeared while we were
        processing it

        * include/ircd_events.h: the pipe for signals is not a socket, so
        we must mark it as such--added SS_NOTSOCK for that special socket;
        events won't be generated if GEN_DESTROY is on, so add GEN_MARKED
        for the benefit of timer_run()

        * configure.in: add --enable-pedantic and --enable-warnings to
        turn on (and off) -Wall -pedantic in CFLAGS

2001-05-21  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_conf.c: change "s_addr" element accesses to "address"
        element accesses

        * include/s_conf.h: on some systems, "s_addr" is a macro; use
        "address" instead

2001-05-18  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/engine_kqueue.c: include ircd_alloc.h; set_or_clear returns
        void in this file; add a missing semi-colon; declare errcode,
        codesize

        * ircd/uping.c (uping_sender_callback): it's pptr, not uping

        * ircd/s_user.c (register_user): comment out spurious reference to
        nextping

        * ircd/s_serv.c (server_estab): comment out spurious reference to
        nextping

        * ircd/s_conf.c (read_configuration_file): comment out spurious
        reference to nextping and nextconnect

        * ircd/s_bsd.c: comment out some spurious references to formerly
        global (now non-existant) variables; correct a couple of typos

        * ircd/s_auth.c: pre-declare some functions referenced in the
        callback; correct a typo

        * ircd/res.c (start_resolver): pass errno value of ENFILE

        * ircd/listener.c (accept_connection): you know your API is messed
        up when...variables that shouldn't have been global crop up in
        other files

        * ircd/list.c (free_client): substitution of == for =

        * ircd/ircd_signal.c: include assert.h for assertion checking;
        check ev_data() to find out what signal generated event

        * ircd/ircd_events.c: some references to the variable "timer"
        should have been references to the variable "ptr"

        * ircd/engine_select.c: it's struct fd_set, not struct fdset;
        ev_timer(ev) is already a timer pointer; declare codesize as a
        size_t to correct signedness issue; use timer_next(), not
        time_next()

        * ircd/engine_poll.c: ev_timer(ev) is already a timer pointer;
        select fd out of struct pollfd in assertion checking; declare
        errcode and codesize; use timer_next(), not time_next()

        * ircd/engine_kqueue.c: ev_timer(ev) is already a timer pointer;
        use function timer_next(), not time_next()

        * ircd/engine_devpoll.c: ev_timer(ev) is already a timer pointer;
        use function timer_next(), not time_next()

        * ircd/Makefile.in (IRCD_SRC): add ircd_events.c to the list of
        compiled sources; do make depend

        * include/list.h: pre-declare struct Connection

        * include/ircd_events.h (gen_ref_inc): cast to the right structure
        name

        * include/s_auth.h: duh; missing */

2001-05-10  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/send.c: update write events status after sending data or
        accumulating data to be sent

        * ircd/m_list.c (m_list): update write events status after
        canceling a running /list

        * ircd/channel.c (list_next_channels): update write events status
        after listing a few channels

        * ircd/s_bsd.c: extensive changes to update to new events model;
        remove on_write_unblocked() and the two implementations of
        read_message(), which have been deprecated by this change

        * ircd/s_auth.c: set the socket events we're interested in for
        clients; simplify some logic that does the connect_nonb followed
        by the socket_add

        * ircd/list.c: define free_connection() to free a connection
        that's become freeable once the struct Socket has been
        deallocated; fix up free_client() to take this new behavior into
        account

        * ircd/ircd.c: call init_client_timer()

        * include/s_bsd.h: declare new REGISTER_ERROR_MESSAGE when unable
        to register connect-in-progress with events system; declare
        init_client_timer() (HACK!) to preserve rate-limiting behavior

        * include/list.h: declare new free_connection()

        * include/client.h: add a struct Socket to struct Connection

2001-05-09  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_signal.c: massage the handlers for SIGHUP, SIGINT, and
        SIGTERM into event callbacks; perform the actions in the
        callbacks, since they're not called in the context of the signal;
        set up the signal callbacks in the event engine

        * ircd/ircd_events.c (signal_callback): we're supposed to look for
        a specific signal; don't generate an event if there is no signal
        structure for it

        * ircd/ircd.c: nuke nextconnect and nextping and replace them with
        connect_timer and ping_timer; massage try_connections() and
        check_pings() into timer callbacks that re-add themselves at the
        right time; remove ircd.c's "event_loop()"; initialize the event
        system and the connect_timer and ping_timer

        * ircd/uping.c: correct a couple more typos

        * ircd/s_auth.c: rework to use new events system

        * ircd/os_solaris.c (os_connect_nonb): update to new interface

        * ircd/os_openbsd.c (os_connect_nonb): update to new interface

        * ircd/os_linux.c (os_connect_nonb): update to new interface

        * ircd/os_generic.c (os_connect_nonb): update to new interface

        * ircd/os_bsd.c (os_connect_nonb): update to new interface

        * include/s_auth.h: remove deprecated members of struct
        AuthRequest, replacing them with struct Socket and struct Timer
        structures; add flags to indicate when these structures have been
        released by the event system; remove the deprecated
        timeout_auth_queries()

        * include/ircd_osdep.h (os_connect_nonb): connect could complete
        immediately, so change the interface to handle that possibility

        * ircd/uping.c (uping_server): noticed and corrected a typo

        * ircd/listener.c: set up to use ircd_event's struct Socket by
        adding an socket_add() call to inetport(), replacing
        free_listener() with socket_del() in close_listener(), and
        reworking accept_connection to be called as the callback

        * ircd/ircd.c: add a call to IPcheck_init()

        * ircd/IPcheck.c: remove IPcheck_expire(); rework
        ip_registry_expire() to be called from a timer; write
        IPcheck_init() to set up the expiration timer (hard-coded for a
        60-second expiration time)

        * include/listener.h: add a struct Socket to the struct Listener;
        remove accept_connection()

        * include/IPcheck.h: add IPcheck_init(), remove IPcheck_expire()

2001-05-08  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_events.c: include config.h; use USE_KQUEUE and
        USE_DEVPOLL instead of HAVE_KQUEUE and HAVE_DEVPOLL_H

        * ircd/engine_select.c: include config.h; set FD_SETSIZE to
        MAXCONNECTIONS, not IRCD_FD_SETSIZE...

        * ircd/engine_poll.c: include config.h

        * ircd/engine_kqueue.c: include config.h

        * ircd/engine_devpoll.c: include config.h

        * ircd/Makefile.in: include engine sources in compilation and make
        depend steps

        * configure.in: add checks for enabling the /dev/poll- and
        kqueue-based engines

        * acconfig.h: add lines for USE_DEVPOLL and USE_KQUEUE

        * ircd/Makefile.in: work in the engine sources

2001-05-07  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_settime.c: include ircd_snprintf.h

        * ircd/ircd_relay.c: stomp a couple of gcc warnings suggesting
        parens around a construct that had both || and &&

        * ircd/chkconf.c: #include "config.h" to get some important
        definitions

        * ircd/Makefile.in: revamp ircd makefile for new build system

        * doc/Makefile.in: revamp doc makefile for new build system

        * config/*: Removed old build system files

        * stamp-h.in: a stamp file

        * install-sh: install-sh for new build system

        * configure.in: configure.in for new build system

        * configure: configure script for new build system (built by
        autoconf)

        * config.sub: config.sub for new build system

        * config.h.in: config.h.in for new build system (built by
        autoheader)

        * config.guess: config.guess for new build system

        * aclocal.m4: aclocal.m4 for new build system (built by aclocal
        1.4)

        * acinclude.m4: aclocal.m4 macros for new build system

        * acconfig.h: config.h skeleton for new build system

        * Makefile.in: modify for new build system

2001-05-01  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_err.c: get rid of the last vestiges of TIME_T_FMT

        * ircd/m_settime.c: get rid of the last vestiges of TIME_T_FMT

        * ircd/m_server.c: get rid of the last vestiges of TIME_T_FMT

2001-05-01  Perry Lorier       <Isomer@coders.net>
        * doc/iauth.doc: Protocol for iauth server. (from hybrid).
        * doc/linux-poll.patch: Patch to make Linux under 2.2 not deadlock
                when you have far far too many sockets in use.
        * {include,ircd}/iauth.c: A start on iauth support.

2001-05-01  Perry Lorier       <Isomer@coders.net>
        * ircd/s_err.c: Suggested wording change.
        * ircd/s_user.c: Users aren't target limited against +k users.
        * ircd/chkconf.c: Made it compile again, who knows if it works, but
                now I can at least make install
        * various: Cleanups on m_*.c files.


2001-04-23  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_misc.c (exit_client): make netsplit server notice say the
        right thing

        * ircd/m_links.c (m_links_redirect): forward-port RPL_ENDOFLINKS
        change to make Khaled happy...

        * ircd/m_whois.c (do_whois): pull-up of m_whois() fix
        (do_whois): duh...

2001-04-21  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/msgq.c: finally remove the msgq_integrity() hack, as it's
        turned up no more bugs

        * ircd/ircd.c: use /* */ comments instead of // comments--all the
        world's not gcc :(

        * ircd/s_conf.c (conf_add_server): use /* */ comments instead of
        // comments--all the world's not gcc :(

        * ircd/runmalloc.c: finally garbage-collect unused file

        * include/runmalloc.h: finally garbage-collect unused file

        * ircd/<multiple files>: addition of '#include "config.h"' before
        all other includes in most .c files

        * include/<multiple files>: remove includes of config.h, which are
        now going into the raw .c files

2001-04-20  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_whois.c (do_whois): display proper server name if the
        user is looking up himself

        * ircd/m_who.c (m_who): disable match by servername or display of
        server names by non-opers

        * include/ircd_policy.h: add define for
        HEAD_IN_SAND_WHO_SERVERNAME to cover full intent of sub-motion 15
        of CFV 165

2001-04-18  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_conf.c: keep the $R in memory so we can see it clearly
        when we do a /stats k

        * ircd/s_user.c (set_user_mode): pull-up of changes to prevent
        users from turning on +s and +g

        * ircd/s_misc.c (exit_client): pull-up of changes to turn off
        net.split notice

        * ircd/parse.c: pull-up of changes to disable /trace, /links, and
        /map for users

        * ircd/m_whois.c (do_whois): pull-up of server name masking for
        /whois

        * ircd/m_user.c (m_user): removal of umode and snomask defaulting
        functions, pull-up

        * ircd/m_stats.c (m_stats): pull-up of stats-disabling stuff

        * ircd/m_map.c (m_map_redirect): pull-up of m_map_redirect()

        * ircd/m_links.c (m_links_redirect): pull-up of m_links_redirect()

        * ircd/channel.c (channel_modes): pull-up of channel key display
        as *

        * include/ircd_policy.h: pull-up of ircd_policy.h

        * include/client.h: pull-up of Set/ClearServNotice()

        * ircd/gline.c (do_gline): report client name in G-line message
        (pull-up)

        * ircd/s_user.c (register_user): pull-up--show IP address in some
        server notices dealing only with users; report which connection
        class has filled up

        * ircd/s_stats.c (report_deny_list): use conf->flags &
        DENY_FLAGS_IP insteaf of conf->ip_kill

        * ircd/m_stats.c (report_klines): use conf->flags & DENY_FLAGS_IP
        insteaf of conf->ip_kill

        * ircd/s_conf.c: use flags field in struct DenyConf; pull-up of
        K-line by real name

        * include/s_conf.h: use a flags field in struct DenyConf; define
        DENY_FLAGS_FILE, DENY_FLAGS_IP, and DENY_FLAGS_REALNAME for
        pull-up of K-line by real name

        * ircd/m_trace.c: pull-up of IP show for user connections

        * doc/example.conf: pull-up of the realname K-line documentation

        * ircd/ircd.c: forward port of pid file advisory locking mechanism

2001-04-16  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/send.c (sendcmdto_flag_butone): recast to just broadcast to
        all servers, rather than to only servers that have +w/+g/whatever
        users on them; among other things, this removes that atrocity
        known as sentalong[] from this function

        * ircd/m_admin.c: must include ircd.h to declare "me"; must
        include hash.h to declare FindUser()

        * ircd/m_wallusers.c: implementation of WALLUSERS

        * ircd/m_desynch.c (ms_desynch): only send DESYNCHs to opers

        * ircd/m_wallops.c: only send WALLOPS to opers

        * ircd/parse.c: add WALLUSERS command to parser table

        * include/handlers.h: declare wallusers handlers

        * include/msg.h: add WALLUSERS command

        * ircd/send.c (sendcmdto_flag_butone): if FLAGS_OPER is or'd with
        flag, send only to appropriate opers

2001-04-13  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/uping.c: refit to use the new events interface

        * ircd/res.c: refit to use the new events interface

        * ircd/ircd_events.c: create timer_chg(), which permits a
        (non-periodic) timer's expire time to be modified; change the
        logic in timer_run() so that timers that were re-added while the
        event was being processed will not be destroyed prematurely

        * include/uping.h: include the events header, declare some extra
        fields in struct UPing, remove timeout value, and define some
        flags for marking which cleanup items have yet to be done

        * include/ircd_events.h: add a prototype for timer_chg() to change
        the expire time of a running timer

2001-03-13 Joseph Bongaarts <foxxe@wtfs.net>
        * ircd/os_openbsd.c: Tweaked the openbsd hack a bit.
        
2001-03-07  Joseph Bongaarts  <foxxe@wtfs.net>

        * config/configure.in: Add check for OpenBSD

        * ircd/os_openbsd.c: Add seperate os dep file for openbsd which
        differs from generic BSD, particularly in its handling of
        _XOPEN_SOURCE.
        
2001-02-12  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_gline.c (ms_gline): propagate a G-line that happened to
        have been added by a U-lined server, rather than going through the
        activate/deactivate logic; propagate G-line removals by U-lined
        servers as well

        * ircd/gline.c: rename propagate_gline() to gline_propagate();
        make gline_propagate() return an int 0 (convenience return); only
        update lastmod in gline_activate() and gline_deactivate() if the
        current lastmod is non-zero, since 0 lastmod is our flag of a
        U-lined server having added a G-line

        * include/gline.h (gline_propagate): exporting the G-line
        propagation function

        * ircd/m_list.c (m_list): duh; permit explicit channel name
        specification only when /list gets two arguments ("Kev
        #wasteland") rather than when /list gets more than two
        arguments--nice braino

2001-01-29  Thomas Helvey <twhelvey1@home.com>

        * ircd/ircd_reply.c (need_more_params): fix bug that allowed
        unregistered clients to spam opers with protocol violation
        messages. Note: the bugfix may have eliminated some useful
        protocol violation messages.
        Please send protocol violation messages explicitly from the
        functions they are discovered in, you have much better context
        for the error there and it helps to document the behavior of the
        server. This was also a design bug in that it violated the
        "A function should do one thing" heuristic. Patching this one
        would have resulted in a continuous spawning of other bugs over
        the next 3 years, so I killed it. Check around for stuff this
        broke and readd the calls to protocol_violation in the functions
        that need to send the message.

2001-01-29  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/channel.c (mode_parse_ban): stopper a tiny leak--if a ban
        already existed, then the logic would (attempt to) skip it, but
        would not free the ban string; now the ban string is free'd and
        the ban count is decremented, releasing the ban for use

        * ircd/s_user.c: make send_umode_out() take a prop argument
        instead of testing for the PRIV_PROPAGATE privilege itself; fix
        set_umode() to use this new argument, calculating it before
        calculating the new privileges for a -o'd user

        * ircd/m_oper.c (m_oper): pass the new prop argument to
        send_umode_out()

        * ircd/channel.c (mode_parse_ban): turn off MODE_ADD bit in bans
        that we're not actually going to add because they already exist;
        test that particular bit before adding to the linked list

        * include/s_user.h: add a prop argument to send_umode_out() to
        indicate whether or not to propagate the user mode

2001-01-24  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/msgq.c: ircd_vsnprintf() returns the number of bytes that
        it would have written; upper-bound the number to prevent overflows
        by proxy; also, tune buffer size given to ircd_vsnprintf() to take
        into account the fact that ircd_vsnprintf() already takes the
        terminal \0 into account

2001-01-22  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/msgq.c: add an incredibly ugly hack to attempt to track
        down an apparent buffer overflow; remove msgq_map(), since it's no
        longer used anywhere; slight tweaks to prevent off-by-one errors,
        but these can't explain the problems we've seen

        * include/msgq.h: remove msgq_map(), since it's no longer used
        anywhere

2001-01-18  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_user.c (set_nick_name): call client_set_privs() after
        parsing user modes

2001-01-17  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_bsd.c (read_message): fix a typo in the select version of
        read_message()

        * ircd/whowas.c (whowas_free): MyFree() is a macro that expects
        its argument to be an lvalue, which means we can't use
        whowas_clean()'s handy-dandy "return ww" feature

        * ircd/ircd_features.c: default LOCOP_KILL to TRUE--oops...

2001-01-16  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_events.c (timer_run): it's possible that the timer got
        deleted during the callback processing, so don't go to the bother
        of requeuing it if the destroy flag is set

        * ircd/engine_select.c: define FD_SETSIZE to be IRCD_FD_SETSIZE
        out of config.h if this is a *BSD; include errno.h (oops);
        decrement error count after an hour using a timer; use FD_SETSIZE
        constant instead of IRCD_FD_SETSIZE constant; fill in event
        processing code

        * ircd/engine_poll.c: include errno.h (oops); decrement error
        count after an hour using a timer; fill in event processing code

        * ircd/engine_kqueue.c: include errno.h (oops); decrement error
        count after an hour using a timer; assert events filter is either
        EVFILT_READ or EVFILT_WRITE; fill in event processing code

        * ircd/engine_devpoll.c: include errno.h (oops); decrement error
        count after an hour using a timer; fill in event processing code

2001-01-15  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/client.c: fixed feattab; basically, when I changed features
        to use small integers specifying bit positions, instead of the
        bits themselves, I forgot to update feattab to not | these
        privileges together; also fixed a bug in the antiprivs masking
        loop in client_set_privs()--last index wouldn't get parsed

2001-01-11  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_events.c: call event_generate() with new data
        argument; make it set that field in struct Event; make
        socket_add() return the value of the eng_add callback

        * ircd/engine_select.c: make engine_add() return a
        successful/unsuccessful status; add bounds-checking outside of an
        assert; use accessor macros; use log_write(), not the deprecated
        ircd_log(); add an assert to engine_loop() to double-check for
        data structure corruption

        * ircd/engine_poll.c: make engine_add() return a
        successful/unsuccessful status; add bounds-checking outside of an
        assert; use accessor macros; use log_write(), not the deprecated
        ircd_log(); add an assert to engine_loop() to double-check for
        data structure corruption

        * ircd/engine_kqueue.c: implementation of an engine for kqueue()

        * ircd/engine_devpoll.c: implementation of an engine for /dev/poll

        * include/ircd_events.h: define some accessor macros; add ev_data
        to struct Event for certain important data--errno values, for
        instance; make EngineAdd callback tell us if it was successful or
        not; add extra argument to event_generate(); make socket_add()
        return the status from EngineAdd

2001-01-10  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_events.c: pass initializer information about how many
        total _filedescriptors_ may be opened at once

        * ircd/ircd.c: use exported "running" instead of unexported
        thisServer.running

        * ircd/engine_select.c: implementation of an event engine based on
        select()

        * ircd/engine_poll.c: implementation of an event engine based on
        poll()

        * include/ircd_events.h: pass the engine initializer an integer
        specifing how many _filedescriptors_ may be opened at once

        * include/ircd.h: running has to be exported for the engine_*
        event loops

2001-01-09  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_events.c: include some needed headers; add some
        comments; make evEngines[] const; bundle sig_sock and sig_fd into
        a struct named sigInfo; rework struct evInfo to have a queue of
        _generators_, and only when threaded; added a gen_init() function
        to centralize generator initialization; fix various compile-time
        errors; rework event_add() for new queueing scheme and checked for
        compile-time errors; add casts where needed; spell evEngines[]
        correctly; make engine_name() return const char*

        * include/ircd_events.h: type EventCallBack depends on struct
        Event, so pre-declare it; put _event_ queue into generators, and
        only when threaded; give engine data a union to store both ints
        and pointers; make engine name a const; fix gen_ref_dec() macro;
        make engine_name() return a const char*

        * ircd/ircd_events.c: gen_dequeue() is now exported, so move it
        down with the non-static functions; modify event_execute() to use
        the new gen_ref_dec() to simplify code; make sure event_generate()
        does not generate new events for generators marked for destruction

        * include/ircd_events.h: the engines, at least, may need to modify
        reference counts to keep generators from going away while
        something still points at them, so add reference counter
        manipulators and export gen_dequeue() for them

        * ircd/ircd_events.c: set up the list of engines to try; set up
        the signal struct Socket; rename netInfo to evInfo; move static
        functions near the beginning of the file; do away with
        signal_signal() (since we no longer keep a signal count ourselves)
        and call event_generate() directly from signal_callback--also
        renamed some functions; allow signal_callback() to read up to
        SIGS_PER_SOCK at once from the signal pipe; add event_init() to
        initialize the entire event system; add event_loop() to call the
        engine's event loop; initialize new struct GenHeader member,
        gh_engdata; remove timer_next(); add socket_add() function to add
        a socket; add socket_del() to mark a socket for deletion; add
        socket_state() to transition a socket between states; add
        socket_events() to set what events we're interested in on the
        socket; add engine_name() to retrieve event engine's name

        * include/ircd_events.h: add engine data field to struct
        GenHeader; rename SOCK_ACTION_REMOVE to SOCK_ACTION_DEL; add a
        note about states vs s_events; remove signal count; fold union
        Generator back into struct Event; remove count members from struct
        Generators; redefine engine callbacks to not take a struct
        Engine*; add explanatory comments to callback definitions; add
        some engine callbacks to handle operations; remove struct Engine
        flag member--can detect single flag from eng_signal member; add
        event_init(), event_loop(), engine_name(), and the socket_*()
        functions; make timer_next() a macro to avoid a function call

2001-01-08  Kevin L. Mitchell  <klmitch@mit.edu>

        * include/ircd_events.h: rename to ircd_events.h, since it handles
        events, not just networking stuff; add signal support; more
        structural rearrangement

        * ircd/ircd_events.c: rename to ircd_events.c, since it handles
        events, not just networking stuff; add signal support; more
        structural rearrangement

2001-01-07  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_network.c: implement timer API; add reference counts
        appropriately

        * include/ircd_network.h: firm up some pieces of the interface;
        split out members everything has into a separate structure; add
        reference counts; add timer API

2001-01-06  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_network.c: static data and event manipulation
        functions for new event processing system

        * include/ircd_network.h: data structures for new event processing
        system

2001-01-03  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/whowas.c: Completely re-did the old allocation scheme by
        turning it into a linked list, permitting the
        NICKNAMEHISTORYLENGTH feature to be changed on the fly

        * ircd/s_debug.c (count_memory): use FEAT_NICKNAMEHISTORYLENGTH
        feature instead of old #define

        * ircd/ircd_features.c: add NICKNAMEHISTORYLENGTH feature as an
        integer feature with a notify callback (whowas_realloc)

        * ircd/client.c (client_set_privs): second memset was supposed to
        be over antiprivs, not privs; thanks, Chris Behrens
        <cbehrens@xo.com> for pointing that out...

        * include/whowas.h: new elements for an extra linked list in
        struct Whowas; a notify function for feature value changes

        * include/ircd_features.h: new feature--FEAT_NICKNAMEHISTORYLENGTH

        * config/config-sh.in: NICKNAMEHISTORYLENGTH is now a feature

2001-01-02  Kevin L. Mitchell  <klmitch@mit.edu>

        * config/config-sh.in: get rid of DEFAULT_LIST_PARAMETER
        compile-time option--now in features subsystem

        * ircd/motd.c (motd_init): rework motd_init() to be called as the
        notify function for MPATH and RPATH features (should probably
        split it up a bit, though...)

        * ircd/m_privs.c (mo_privs): if called with no parameters, return
        privs of the caller, rather than an error

        * ircd/m_list.c: pull usage message into its own function; pull
        list parameter processing into its own function that does not
        modify the contents of the parameter; add list_set_default() to
        set the default list parameter (uses the notify hook); rework
        m_list() to make use of these functions; removed dead code

        * ircd/ircd_log.c (log_feature_mark): make sure to return 0, since
        we have no notify handler

        * ircd/ircd_features.c: add notify callback for notification of
        value changes; give mark callback an int return value to indicate
        whether or not to call the notify callback; fix up feature macros
        for new notify callback; add DEFAULT_LIST_PARAM feature; rewrite
        string handling in feature_set() to deal with def_str being a null
        pointer; wrote feature_init() to set up all defaults appropriately

        * ircd/ircd.c (main): call feature_init() instead of
        feature_mark(), to avoid calling notify functions while setting up
        defaults

        * ircd/client.c: updated to deal with new privileges structure

        * ircd/class.c: updated so init_class() can be called should one
        of PINGFREQUENCY, CONNECTFREQUENCY, MAXIMUM_LINKS, or
        DEFAULTMAXSENDQLENGTH be changed

        * include/ircd_log.h: log_feature_mark() updated to fit with new
        API changes

        * include/ircd_features.h: added DEFAULT_LIST_PARAM feature and
        feature_init() function (found necessary since adding the notify
        stuff and notifying motd.c during start-up...before we defined
        RPATH!)

        * include/client.h: move privs around to enable addition of more
        bits if necessary; based on the FD_* macros

        * include/channel.h: declare list_set_default (actually located in
        m_list.c *blanche*)

        * ircd/s_user.c: retrieve MAXSILES and MAXSILELENGTH (now
        AVBANLEN*MAXSILES) from features subsystem

        * ircd/s_debug.c (debug_serveropts): CMDLINE_CONFIG doesn't go to
        anything anymore

        * ircd/s_bsd.c: retrieve HANGONGOODLINK and HANGONRETRYDELAY from
        the features subsystem

        * ircd/s_auth.c (start_auth): NODNS migrated to the features
        subsystem

        * ircd/random.c: created random_seed_set() function to set seed
        value, along with some stuff to make ircrandom() a little more
        random--state preserving, xor of time instead of direct usage,
        etc.; it's still a pseudo-random number generator, though, and
        hopefully I haven't broken the randomness

        * ircd/m_version.c: FEATUREVALUES makes use of feature_int() calls

        * ircd/m_join.c: use features interface to retrieve
        MAXCHANNELSPERUSER

        * ircd/ircd_features.c: add NODISP flag for super-secret features;
        add a whole bunch of new features migrated over from make config

        * ircd/ircd.c: use features interface to retrieve PINGFREQUENCY,
        CONNECTTIMEOUT, and TIMESEC

        * ircd/client.c (client_get_ping): use features interface to
        retrieve PINGFREQUENCY

        * ircd/class.c: use features interface to retrieve PINGFREQUENCY,
        CONNECTFREQUENCY, MAXIMUM_LINKS, and DEFAULTMAXSENDQLENGTH

        * ircd/chkconf.c (DEFAULTMAXSENDQLENGTH): since it's now in the
        features subsystem, we have to add something explicit

        * ircd/channel.c: use features interface to retrieve
        KILLCHASETIMELIMIT, MAXBANLENGTH, MAXBANS, and MAXCHANNELSPERUSER;
        note that MAXBANLENGTH is now calculated dynamically from MAXBANS
        and AVBANLEN

        * ircd/Makefile.in: run make depend

        * include/supported.h (FEATURESVALUES): update to reference
        feature settings

        * include/random.h: add prototype for random_seed_set

        * include/ircd_features.h: add several more features

        * include/channel.h: move MAXBANS and MAXBANLENGTH into feature
        subsystem

        * config/config-sh.in: feature-ized some more stuff

        * include/motd.h: some new elements in motd.h for motd.c changes

        * ircd/motd.c: motd_cache() now searches a list of already cached
        MOTD files; saves us from having duplicate caches in memory if
        there are two identical T-lines for two different sites...

2001-01-02  Perry Lorier <isomer@coders.net>
        * ircd/motd.c: don't core if the motd isn't found.  Bug found by
        Amarande.

2001-01-02  Perry Lorier <isomer@coders.net>
        * ircd/s_err.c: Added third param to 004 - the channel modes that tage params.  Used by hybrid/epic.
        * ircd/s_channels.c: Added fix for msg'ing a -n+m channel - thanks
                to guppy for noticing, and hektik for providing the fix.
        * misc others: Minor cleanups, added more protocol_violations, ripped
                out more P09 stuffs, bit more protocol neg stuff.

2000-12-19  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_ison.c (m_ison): Dianora says that ISON has to end with a
        space (*sigh* stupid clients...)

        * ircd/s_user.c: make WALLOPS_OPER_ONLY a feature managed through
        ircd_features.[ch]

        * ircd/s_err.c: get rid of GODMODE conditionals

        * ircd/s_debug.c (debug_serveropts): switch to using appropriate
        calls into the features subsystem for various serveropts
        characters

        * ircd/s_conf.c (find_conf_entry): get rid of USEONE conditional

        * ircd/s_bsd.c: remove GODMODE conditional; use features subsystem
        to get value of VIRTUAL_HOST and CLIENT_FLOOD; remove
        NOFLOWCONTROL conditional

        * ircd/s_auth.c: use features subsystem to determine value of
        KILL_IPMISMATCH

        * ircd/parse.c: get rid of NOOPER and GODMODE conditionals; use
        features subsystem to determine the setting of IDLE_FROM_MSG

        * ircd/numnicks.c: get rid of EXTENDED_NUMERICS conditionals

        * ircd/motd.c: get value of NODEFAULTMOTD from features subsystem;
        use features subsystem to get motd file names

        * ircd/m_settime.c: get value of RELIABLE_CLOCK from features
        subsystem

        * ircd/m_server.c: get rid of CRYPT_LINK_PASSWORD, since it does
        us no good; use features subsystem to figure out if we need to do
        HUB-type stuff; make TESTNET debugging sendto_opmask_butone's use
        the Debug(()) macro instead; get value of RELIABLE_CLOCK from
        features subsystem

        * ircd/m_privmsg.c: get IDLE_FROM_MSG from the features subsystem

        * ircd/m_oper.c: get CRYPT_OPER_PASSWORD from the features
        subsystem

        * ircd/m_connect.c: get SERVER_PORT from the features subsystem

        * ircd/ircd_log.c (log_set_file): fix a bug that kept log files
        from getting marked if they were already set to something...

        * ircd/ircd_features.c: add a flag to indicates read-only access;
        add several new features that used to be compile-time selected

        * ircd/ircd.c: grab pidfile out of feature subsystem; don't check
        access to motd files (what the heck?); make sure to initialize the
        feature subsystem before trying to write the config file

        * ircd/dbuf.c: use feature_int() to retrieve BUFFERPOOL settings;
        use feature_bool() to figure out if we're using the FERGUSON
        flusher

        * ircd/Makefile.in: MPATH and RPATH are now done differently, so
        remove the clause that creates empty files of that name; also ran
        make depend

        * include/sys.h: CLIENT_FLOOD is now a feature; unfortunately,
        there is no easy way to bounds-check it at present

        * include/querycmds.h: make sure ircd_features.h is included; use
        feature_str(FEAT_DOMAINNAME) in calls to match()

        * include/ircd_features.h: many new features that used to be
        compile-time selected

        * config/config-sh.in: add * to DOMAINNAME; try also using first
        argument to search in /etc/resolv.conf; removed many compile-time
        options that now can be configured through the features system

2000-12-18  Kevin L. Mitchell  <klmitch@mit.edu>

        * doc/api/log.txt: how to use the logging API

        * doc/api/features.txt: how to use the features API

        * doc/api/api.txt: how to write API documentation

        * include/ircd_features.h: rearranged a couple of features for
        neatness purposes

        * ircd/ircd_features.c: cleaned up the macros some; rearranged
        some code to all go into the switch; rearranged a couple of
        features for neatness purposes

2000-12-16  Greg Sikorski <gte@atomicrevs.demon.co.uk>
        * ircd/os_bsd.c: Added os_set_tos for BSD users.

2000-12-16  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_features.c: Isomer almost got it right; you need to
        use F_I(), since it's an integer value, not a boolean value.  The
        asserts in feature_int would catch you out...  Also made the F_*
        macros take flags

        * ircd/s_err.c: define RPL_PRIVS reply

        * ircd/parse.c: put new PRIVS command into command table

        * ircd/m_privs.c (mo_privs): message handler to report operator
        privileges

        * ircd/ircd_features.c: declare new features OPER_SET and
        LOCOP_SET; redo boolean testing routine to accept TRUE, YES, and
        ON for boolean TRUE, and FALSE, NO, and OFF for boolean FALSE

        * ircd/client.c: simplify client_set_privs() with a table that
        defines what features to test for; add new client_report_privs()

        * ircd/Makefile.in: compile new m_privs.c; run make depend

        * include/numeric.h (RPL_PRIVS): new reply numeric for displaying
        an operator's privileges

        * include/msg.h: define new command: PRIVS

        * include/ircd_features.h: create new features OPER_SET and
        LOCOP_SET for controlling access to /set

        * include/handlers.h (mo_privs): declare message handler for
        reporting oper privileges

        * include/client.h (client_report_privs): declare function to
        report what privileges an oper has

        * ircd/m_whois.c (do_whois): fix a bug that caused /whois to
        report that a user is an oper if the oper doing the /whois had
        PRIV_SEE_OPERS

2000-12-17  Isomer <Isomer@coders.net>
        * ircd/listener.c: added support for TOS twiddling as a 'feature'.

2000-12-17  Isomer <Isomer@coders.net>
        * ircd/os_linux.c: add TOS stuffs

        * ircd/listener.c: add TOS stuffs

2000-12-16  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/whocmds.c (do_who): use HasPriv to determine whether or not
        to indicate a user is an oper

        * ircd/s_user.c: clear privileges setting when deopping; don't
        propagate +o unless user has PRIV_PROPAGATE privilege

        * ircd/s_debug.c (debug_serveropts): created debug_serveropts()
        function and replaced how the server option string is generated

        * ircd/parse.c: remove conditional on CONFIG_OPERCMDS

        * ircd/m_whois.c (do_whois): use HasPriv to determine whether or
        not to indicate the user is an operator

        * ircd/m_who.c: use HasPriv to determine whether or not a user
        should be displayed in the list of opers

        * ircd/m_version.c: call debug_serveropts() to get server option
        string

        * ircd/m_userip.c (userip_formatter): use HasPriv to determine
        whether or not to show oper status

        * ircd/m_userhost.c (userhost_formatter): use HasPriv to determine
        whether or not to show oper status

        * ircd/m_restart.c (mo_restart): replace ugly #ifdef conditional
        checks with HasPriv check; remove dead code

        * ircd/m_rehash.c (mo_rehash): replace ugly #ifdef conditional
        checks with HasPriv check

        * ircd/m_opmode.c (mo_opmode): use HasPriv to check permissions;
        use feature_bool to check if disabled

        * ircd/m_oper.c (m_oper): set oper priviliges

        * ircd/m_mode.c (m_mode): replace #ifdef conditional with HasPriv
        check

        * ircd/m_kill.c (mo_kill): use HasPriv checks to determine if we
        can kill

        * ircd/m_kick.c (m_kick): replace #ifdef conditional with HasPriv
        check

        * ircd/m_jupe.c (mo_jupe): rework permissions checking structure;
        use feature_bool to check if disabled

        * ircd/m_join.c (m_join): remove BADCHAN conditional; replace
        #ifdef conditional with a HasPriv check

        * ircd/m_gline.c (mo_gline): rework permissions checking
        structure; use feature_bool to check if any part is disabled

        * ircd/m_die.c: replace ugly #ifdef conditionals with HasPriv
        check; remove dead code

        * ircd/m_clearmode.c: use feature_bool() to detect if we're
        disabled; use HasPriv to figure out what we're permitted to do;
        only allow clearmode on moded channels

        * ircd/ircd_features.c: define various features; use HasPriv to
        verify permissions to set/reset

        * ircd/gline.c (gline_add): use HasPriv instead of #ifdef
        conditionals

        * ircd/client.c (client_set_privs): function to set an oper's
        privileges

        * ircd/channel.c: use HasPriv calls instead of #ifdef conditionals

        * include/whocmds.h: deconditionalize several macros and
        substitute appropriate calls to HasPriv()

        * include/s_debug.h: get rid of global serveropts[]; define new
        function debug_serveropts() to build that string on the fly

        * include/ircd_features.h: define some features

        * include/client.h: add privs member to struct Connection; define
        various priviledges

        * include/channel.h: no longer using IsOperOnLocalChannel; remove
        conditional of MAGIC_OPER_OVERRIDE on OPER_WALK_THROUGH_LMODES

        * doc/Configure.help: remove help information for deprecated
        options

        * config/config-sh.in: remove certain deprecated options having to
        do with what opers can and cannot do--first stage in moving
        compile-time constants into the .conf

2000-12-16  Isomer <Isomer@coders.net>
        * ircd/parse.c: detect if the prefix is missing and try and recover
        instead of coring.

2000-12-15  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_log.c: found and fixed some bugs in the debug logging
        code that would sometimes result in the log file not being
        reopened--which meant that a user could connect and get the
        logging output--oops

        * ircd/Makefile.in: run make depend...

        * ircd/s_stats.c: get rid of report_feature_list()

        * ircd/s_err.c: add the 'bad value' error message, shift error
        messages over somewhat

        * ircd/s_debug.c (debug_init): call log_debug_init with the
        use_tty flag

        * ircd/s_conf.c (read_configuration_file): unmark features before
        reading the config file, then reset unmarked features after
        reading the config file

        * ircd/m_stats.c: use feature_report() instead of
        report_feature_list()

        * ircd/ircd_log.c: fix log_debug_file (bogus assertion); add
        special 'mark' flags and use them; add the stuff needed by the
        features API

        * ircd/ircd_features.c: rework the features API and add gobs of
        comments to try to explain what some of these complex functions
        are actually doing

        * include/s_stats.h: get rid of report_feature_list(); use
        feature_report() instead

        * include/numeric.h: added a new error message and shifted old
        values over some--this is, after all, an alpha

        * include/ircd_log.h: log_debug_init now takes an integer to tell
        it if it should be using the tty; added a couple of functions
        required by the features API

        * include/ircd_features.h: add an enum and some more functions to
        flesh out the feature API--it should now be possible to put all
        those compile-time constants in the config file!

        * ircd/send.c: got the direction of the assert incorrect...

        * ircd/send.c: implement the efficiency of flush_connections by
        creating a linked list of struct Connection's with queued data;
        also get rid of flush_sendq_except and make sure to yank
        connections out of the list when their sendQs become empty (notice
        the assertion in flush_connections!)

        * ircd/s_bsd.c (close_connection): must yank the Connection out of
        the sendq list

        * ircd/list.c (dealloc_connection): must yank the Connection out
        of the sendq list

        * ircd/dbuf.c (dbuf_put): call flush_connections instead of the
        deprecated flush_sendq_except

        * ircd/client.c: define a couple new helper functions for sendq
        threading--this will make the flush_connections function in send.c
        considerably more efficient by creating a linked list of
        Connections that have queued data to send

        * include/send.h: remove flush_sendq_except, as it's not used
        anymore

        * include/client.h: declare a couple new helper functions for the
        sendq threading system

2000-12-14  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_ison.c (m_ison): Apply Diane Bruce's patch to make ISON
        parse all arguments

        * ircd/s_debug.c (count_memory): modify to report for clients and
        connections, not local clients and remote clients

        * ircd/list.c: fiddle with the client-fiddling functions to take
        into account the divorce of struct Connection from struct Client

        * ircd/ircd.c: define a struct Connection for me, initialize it,
        and link it into the right place (ewww, globals!)

        * include/client.h: remove CLIENT_{LOCAL,REMOTE}_SIZE; split
        struct Client into struct Client and struct Connection; redefine
        local-portion accessor macros to go through struct Client to the
        struct Connection; define struct Connection accessor macros

2000-12-13  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/whowas.c: missed a couple of accesses to a struct Client

        * ircd/uping.c: missed a couple of accesses to a struct Client

        * ircd/send.c: missed a couple of accesses to a struct Client

        * ircd/s_user.c: missed a couple of accesses to a struct Client

        * ircd/s_misc.c: missed a couple of accesses to a struct Client

        * ircd/s_conf.c: missed a couple of accesses to a struct Client

        * ircd/s_bsd.c: missed a couple of accesses to a struct Client

        * ircd/s_auth.c: missed a couple of accesses to a struct Client

        * ircd/res.c: missed a couple of accesses to a struct Client

        * ircd/parse.c: missed a couple of accesses to a struct Client

        * ircd/m_whois.c: use new accessor macros for struct Client

        * ircd/m_who.c: use new accessor macros for struct Client

        * ircd/m_wallchops.c: use new accessor macros for struct Client

        * ircd/m_version.c: use new accessor macros for struct Client

        * ircd/m_userip.c: use new accessor macros for struct Client

        * ircd/m_userhost.c: use new accessor macros for struct Client

        * ircd/m_user.c: use new accessor macros for struct Client

        * ircd/m_uping.c: use new accessor macros for struct Client

        * ircd/m_trace.c: use new accessor macros for struct Client

        * ircd/m_topic.c: use new accessor macros for struct Client

        * ircd/m_time.c: use new accessor macros for struct Client

        * ircd/m_stats.c: use new accessor macros for struct Client

        * ircd/m_squit.c: use new accessor macros for struct Client

        * ircd/m_silence.c: use new accessor macros for struct Client

        * ircd/m_server.c: use new accessor macros for struct Client;
        remove dead code

        * ircd/m_rpong.c: use new accessor macros for struct Client

        * ircd/m_rping.c: use new accessor macros for struct Client

        * ircd/m_quit.c: use new accessor macros for struct Client

        * ircd/m_privmsg.c: use new accessor macros for struct Client

        * ircd/m_pong.c: use new accessor macros for struct Client; remove
        dead code

        * ircd/m_ping.c: use new accessor macros for struct Client

        * ircd/m_pass.c: use new accessor macros for struct Client

        * ircd/m_part.c: use new accessor macros for struct Client

        * ircd/m_oper.c: use new accessor macros for struct Client

        * ircd/m_notice.c: use new accessor macros for struct Client

        * ircd/m_nick.c: use new accessor macros for struct Client

        * ircd/m_names.c: use new accessor macros for struct Client

        * ircd/m_mode.c: use new accessor macros for struct Client

        * ircd/m_map.c: use new accessor macros for struct Client

        * ircd/m_list.c: use new accessor macros for struct Client

        * ircd/m_links.c: use new accessor macros for struct Client;
        remove some dead code

        * ircd/m_kill.c: use new accessor macros for struct Client; remove
        some dead code

        * ircd/m_kick.c: use new accessor macros for struct Client

        * ircd/m_join.c: use new accessor macros for struct Client; remove
        some dead code

        * ircd/m_ison.c: use new accessor macros for struct Client

        * ircd/m_invite.c: use new accessor macros for struct Client

        * ircd/m_info.c: use new accessor macros for struct Client

        * ircd/m_gline.c: use new accessor macros for struct Client

        * ircd/m_error.c: use new accessor macros for struct Client

        * ircd/m_create.c: use new accessor macros for struct Client

        * ircd/m_connect.c: use new accessor macros for struct Client;
        removed some dead code

        * ircd/m_burst.c: use new accessor macros for struct Client

        * ircd/m_away.c: use new accessor macros for struct Client

        * ircd/m_admin.c: use new accessor macros for struct Client

        * ircd/hash.c: missed a couple of accesses to a struct Client

        * ircd/gline.c: missed a couple of accesses to a struct Client

        * ircd/crule.c: missed a couple of accesses to a struct Client

        * ircd/class.c: missed an access to a struct Client

        * ircd/channel.c: missed a couple of accesses to a struct Client

        * ircd/IPcheck.c: missed an access to a struct Client

        * include/querycmds.h: fix a couple of stats macros to use
        structure accessor macros

        * include/client.h: change structure member names to highlight any
        places in the code I've missed

2000-12-12  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/whowas.c: use new struct Client accessor macros

        * ircd/whocmds.c: use new struct Client accessor macros

        * ircd/send.c: use new struct Client accessor macros

        * ircd/s_user.c: use new struct Client accessor macros; removed
        some dead code

        * ircd/s_serv.c: use new struct Client accessor macros; removed
        some dead code

        * ircd/s_numeric.c: use new struct Client accessor macros

        * ircd/s_misc.c: use new struct Client accessor macros

        * ircd/s_debug.c: use new struct Client accessor macros

        * ircd/s_conf.c: use new struct Client accessor macros

        * ircd/s_bsd.c: use new struct Client accessor macros

        * ircd/s_auth.c: use new struct Client accessor macros

        * ircd/parse.c: use new struct Client accessor macros

        * ircd/packet.c: use new struct Client accessor macros

        * ircd/numnicks.c: use new struct Client accessor macros

        * ircd/motd.c: use new struct Client accessor macros

        * ircd/listener.c: use new struct Client accessor macros

        * ircd/list.c: use new struct Client accessor macros

        * ircd/jupe.c: use new struct Client accessor macros

        * ircd/ircd_snprintf.c: use new struct Client accessor macros

        * ircd/ircd_reply.c: use new struct Client accessor macros

        * ircd/ircd_relay.c: use new struct Client accessor macros

        * ircd/ircd.c: use new struct Client accessor macros

        * ircd/gline.c: catch some instances of me.<stuff> I missed
        previously

        * ircd/client.c: use cli_ instead of con_

        * ircd/class.c: use cli_ instead of con_

        * ircd/channel.c: use cli_ instead of con_

        * ircd/IPcheck.c: use cli_ instead of con_; catch some instances
        of me.<stuff> I missed previously

        * include/client.h: use cli_ instead of con_...seemed like a good
        idea at the time *shrug*

2000-12-11  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/hash.c: use struct Client accessor macros

        * ircd/gline.c: use struct Client accessor macros

        * ircd/crule.c: use struct Client accessor macros

        * ircd/client.c: use struct Client accessor macros; remove some
        dead code

        * ircd/class.c: use struct Client accessor macros

        * ircd/channel.c: use struct Client accessor macros; remove some
        dead code

        * ircd/IPcheck.c: use struct Client accessor macros

        * include/numnicks.h: use struct Client accessor macros

        * include/client.h: first step to divorcing struct Client and
        struct Connection--define accessor macros and use them

        * ircd/gline.c: When Uworld removed Uworld-set G-lines, only the
        uplink would remove them.  This is because the removal protocol
        message wasn't being sent to the uplinks.  This is fixed by fixing
        propagate_gline() to send the proper number of arguments depending
        on whether or not we're adding or deleting the Uworld gline, and
        by having gline_deactivate() make sure to turn off the active bit
        and call propagate_gline() if it's a Uworld gline

2000-12-10  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/os_generic.c: make sure IOV_MAX gets defined, just in case

        * ircd/os_bsd.c: apparently BSD doesn't have IOV_MAX defined
        anywhere intelligent...

2000-12-09  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/send.c (send_queued): call deliver_it with appropriate
        arguments

        * ircd/s_serv.c: reorder a couple of headers--cosmetic

        * ircd/s_bsd.c (deliver_it): make deliver_it work with a struct
        MsgQ

        * ircd/os_solaris.c (os_sendv_nonb): function for calling writev
        with appropriate iovec

        * ircd/os_linux.c (os_sendv_nonb): function for calling writev
        with appropriate iovec

        * ircd/os_generic.c (os_sendv_nonb): function for calling writev
        with appropriate iovec

        * ircd/os_bsd.c (os_sendv_nonb): function for calling writev with
        appropriate iovec

        * ircd/msgq.c (msgq_mapiov): add a len_p argument for totalling up
        exactly how much we're trying to write out to the fd

        * include/s_bsd.h: make deliver_it take a struct MsgQ

        * include/msgq.h: add a len_p argument to msgq_mapiov to help
        detect short writes that indicate possible socket blocking

        * include/ircd_osdep.h: declare os_sendv_nonb()

        * ircd/channel.c (modebuf_mode): don't add empty modes...

2000-12-08  Kevin L. Mitchell  <klmitch@mit.edu>

        * include/send.h: add prio argument to send_buffer to select
        between normal and priority queues

        * ircd/s_user.c (send_user_info): add prio argument to send_buffer
        call

        * ircd/m_ison.c (m_ison): add prio argument to send_buffer call

        * ircd/ircd_reply.c (send_reply): add prio argument to send_buffer
        call

        * ircd/channel.c (send_channel_modes): add prio argument to
        send_buffer call

        * ircd/send.c (send_buffer): add a prio argument to select the
        priority queue; update send.c functions to use it

        * ircd/msgq.c (msgq_add): remove msgq_prio; fold msgq_link and
        msgq_add; add a prio argument to msgq_add to select the priority
        queue

        * include/msgq.h: remove msgq_prio; add a prio argument to
        msgq_add

        * ircd/send.c: remove sendbuf; remove GODMODE code; switch to
        using msgq functions instead of dbuf functions; remove old, dead
        sendto_* functions; redo send_buffer to take a struct MsgBuf;
        rework sendcmdto_* functions to make use of the new struct MsgBuf

        * ircd/s_user.c: remove hunt_server; restructure send_user_info to
        make appropriate use of struct MsgBuf

        * ircd/s_debug.c (count_memory): count memory used by the MsgQ
        system and report it

        * ircd/s_conf.c (read_configuration_file): use
        sendto_opmask_butone instead of the now dead sendto_op_mask

        * ircd/s_bsd.c: switch to using appropriate MsgQLength and other
        calls on sendQ

        * ircd/parse.c (parse_server): get rid of a piece of GODMODE code

        * ircd/msgq.c: add msgq_append and msgq_bufleft; fix a bug in
        msgq_clean

        * ircd/m_version.c: fix spelling in comments marking dead code

        * ircd/m_userip.c (userip_formatter): restructure to make use of
        struct MsgBuf

        * ircd/m_userhost.c (userhost_formatter): restructure to make use
        of struct MsgBuf

        * ircd/m_stats.c: use MsgQLength on a sendQ

        * ircd/m_settime.c: use MsgQLength instead of DBufLength on a
        sendQ; mark a piece of dead code

        * ircd/m_names.c: use send_reply instead of sendto_one

        * ircd/m_mode.c: use new mode; remove old dead code

        * ircd/m_ison.c (m_ison): restructure to make use of struct MsgBuf

        * ircd/m_burst.c: use BUFSIZE instead of IRC_BUFSIZE; remove old
        dead code

        * ircd/listener.c (accept_connection): use sendto_opmask_butone
        instead of sendto_op_mask

        * ircd/list.c (free_client): use MsgQClear to clear sendQ

        * ircd/ircd_reply.c: remove send_error_to_client; restructure
        send_reply to make use of struct MsgBuf

        * ircd/dbuf.c (dbuf_put): remove argument to flush_sendq_except,
        since its no longer used (at least currently)

        * ircd/channel.c: restructure send_channel_modes to make use of
        struct MsgBuf; remove set_mode, add_token_to_sendbuf, cancel_mode,
        and send_hack_notice; use BUFSIZE instead of IRC_BUFSIZE

        * ircd/Makefile.in: add msgq.c to list of sources; run make depend

        * ircd/IPcheck.c: use sendcmdto_one instead of sendto_one

        * include/send.h: send_buffer now takes a struct MsgBuf * instead
        of a char *; flush_sendq_except now takes no arguments, as sendq
        flushing currently only happens in dbuf.h and sendQ is a struct
        MsgQ; remove prototypes for a lot of old sendto_* functions that
        aren't used anymore; remove sendbuf and IRC_BUFSIZE--the former is
        no longer needed, and the latter is identical to BUFSIZE in
        ircd_defs.h

        * include/s_user.h: make InfoFormatter take a struct MsgBuf*
        instead of a char *; also make it return void, instead of char *

        * include/msgq.h: add msgq_append and msgq_bufleft functions

        * include/client.h: use a struct MsgQ instead of a struct DBuf for
        sendq

        * doc/Configure.help: Remove help for compile-time options that
        have gone away

        * config/config-sh.in: remove CONFIG_NEWMODE

        * ircd/m_server.c (mr_server): don't send server IPs in any server
        notices

        * ircd/msgq.c (msgq_vmake): add \r\n to messages

2000-12-07  Kevin L. Mitchell  <klmitch@mit.edu>

        * include/msgq.h: declare the MsgQ API

        * ircd/msgq.c: implementation of new MsgQ system

2000-12-06  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_features.c: #include was supposed to be for
          ircd_features.h, not features.h--missed when I had to do a
          rename because of namespace collision

2000-12-05  Greg Sikorski <gte@atomicrevs.demon.co.uk>
        * ircd/m_topic.c: Added missing braces that caused all remote
          topics to be ignored.

2000-12-04  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_create.c: I'm tired of the exit_client warning :)
        (ms_create): discovered that exit_client() was being called with
        too few arguments

        * ircd/s_misc.c (exit_client): remove all dependance on
        FNAME_USERLOG, since that's now gone; log only to LS_USER

        * ircd/s_debug.c: USE_SYSLOG no longer means anything

        * ircd/m_oper.c (m_oper): no longer log to LS_OPERLOG--we already
        log to LS_OPER

        * ircd/m_kill.c: no longer conditionalize on SYSLOG_KILL

        * ircd/ircd_log.c: remove LS_OPERLOG, LS_USERLOG

        * include/ircd_log.h: remove LS_OPERLOG, LS_USERLOG--they serve
        the same purpose as LS_USER and LS_OPER

        * config/config-sh.in: remove no longer relevant log config
        variables

        * ircd/uping.c (uping_init): use log_write instead of ircd_log

        * ircd/s_misc.c (exit_client): use log_write instead of ircd_log

        * ircd/s_conf.c: use log_write instead of ircd_log

        * ircd/s_bsd.c (report_error): use log_write instead of ircd_log

        * ircd/s_auth.c (timeout_auth_queries): use log_write instead of
        ircd_log

        * ircd/res.c (send_res_msg): use log_write instead of ircd_log

        * ircd/m_who.c: use log_write instead of write_log; no longer
        conditionalize on WPATH; mark dead ircd_log calls

        * ircd/m_uping.c: mark dead ircd_log call

        * ircd/m_server.c (mr_server): use log_write instead of ircd_log

        * ircd/m_restart.c: use log_write instead of ircd_log; mark dead
        ircd_log calls

        * ircd/m_rehash.c (mo_rehash): use log_write instead of ircd_log

        * ircd/m_oper.c: use log_write instead of ircd_log; no longer
        conditionalize on FNAME_OPERLOG; mark dead ircd_log calls

        * ircd/m_kill.c: mark dead ircd_log calls

        * ircd/m_connect.c: use log_write instead of ircd_log; mark dead
        ircd_log

        * ircd/m_clearmode.c: use log_write instead of write_log; no
        longer conditionalize on OPATH

        * ircd/jupe.c: use log_write instead of write_log; no longer
        conditionalize on JPATH

        * ircd/ircd_log.c: add USER subsystem; remove ircd_log() compat
        function; fix a couple of bugs

        * ircd/ircd_alloc.c: fixed a comment

        * ircd/ircd.c: use log_write instead of ircd_log; fold server
        notice generation in a couple of cases

        * ircd/gline.c: use log_write instead of write_log; no longer
        conditionalize on GPATH

        * ircd/channel.c (modebuf_flush_int): use log_write instead of
        write_log; no longer conditionalize on OPATH

        * ircd/Makefile.in: run make depend, since dependencies have
        changed

        * doc/example.conf: add system USER to documentation

        * include/ircd_log.h: add system USER; remove old ircd_log()
        declarations

2000-12-04  Isomer <isomer@coders.net>
        * ircd/m_names.c: Add NAMES_EON to do_names to say add a
        'end_of_names' reply when done.
        * ircd/m_join.c: use NAMES_EON as mentioned above

2000-12-01  net  <simms@LUCIDA.QC.CA>

        * ircd/motd.c: add a freelist for struct Motds

2000-11-30  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_stats.c (report_feature_list): report features--only
        local opers can see logging configuration, since it doesn't really
        mean anything to users

        * ircd/s_err.c: add reply messages for new feature subsystem

        * ircd/s_conf.c: add F lines to .conf

        * ircd/parse.c: add the message descriptions for /set, /reset, and
        /get

        * ircd/m_stats.c: add /stats f

        * ircd/m_set.c (mo_set): implement /set

        * ircd/m_reset.c (mo_reset): implement /reset

        * ircd/m_rehash.c: /rehash m now flushes MOTD cache, and /rehash l
        reopens log files (for log file rotation)

        * ircd/m_get.c (mo_get): implement /get

        * ircd/ircd_log.c: use int instead of void return value; add
        log_report_features() and log_canon(); fix a function that
        disappears if DEBUGMODE not #define'd

        * ircd/ircd_features.c: functions to manipulate feature settings
        either from the config file or with the new /set, /reset, and /get
        commands

        * ircd/Makefile.in: add new .c files, run make depend

        * include/s_stats.h: declare report_feature_list() (/stats f
        handler)

        * include/numeric.h: add RPL_STATSFLINE, RPL_FEATURE,
        ERR_NOFEATURE, ERR_BADLOGTYPE, ERR_BADLOGSYS, and ERR_BADLOGVALUE
        reply numerics

        * include/msg.h: add defines for SET, RESET, and GET

        * include/ircd_log.h: add a function to canonicalize subsystem
        names; change some void return values to int

        * include/ircd_features.h: new features subsystem handles all the
        manipulation of special features, like log files

        * include/handlers.h: declare new mo_{s,res,g}et message handlers
        for fiddling with features run-time

        * include/client.h (SNO_DEFAULT): don't set SNO_DEBUG by default;
        seemed like a good idea at the time...

        * doc/example.conf: document new F lines

2000-11-29  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_debug.c: rewrite debug_init() and vdebug() in terms of
        new logging functions, which have special support for the debug
        log; added loop detection to vdebug(), so that I can
        sendto_opmask_butone() from log_vwrite() without incurring another
        call to vdebug()

        * ircd/s_conf.c (rehash): call log_reopen() from rehash routine;
        this allows log file rotations

        * ircd/m_kill.c: call log_write_kill() instead of ircd_log_kill()

        * ircd/ircd_log.c: much more work fleshing out the interface;
        removed old interface; included backwards-compat ircd_log()
        function that logs to subsystem LS_OLDLOG

        * ircd/ircd.c: switch to new log_init()/log_close()/log_reopen()
        functions

        * include/ircd_log.h: include stdarg.h for va_list; move ordering
        warning to top of file; fill out LogSys enum; declare new
        log_debug_init(), log_vwrite(), log_write_kill(), and
        log_[sg]et_*() functions; add flags argument to log_write();
        defined flags to inhibit various logging actions

        * include/client.h: added support for new SNO_DEBUG, enabled only
        if DEBUGMODE is defined

2000-11-28  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_log.c: make sure the various LOG_* constants are
        defined (probably not needed, since #include <syslog.h> isn't
        conditional); various static data needed for the new logging
        functions; definitions of new logging functions

        * include/ircd_log.h: new LogSys enum, declarations for part of
        new logging API

        * ircd/motd.c: we were setting type to MOTD_CLASS unconditionally,
        which was of course stupid; switched to using switch/case in
        initialization in motd_create(); zero the MotdList.other pointer
        from motd_clear()

        * ircd/ircd.c (main): motd_init() has to come before init_conf(),
        or we overwrite init_conf()'s hard work with respect to T-lines

2000-11-27  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_stats.c: comment out report_motd_list and include a
        reference to motd_report()

        * ircd/s_conf.c: rip out the old MOTD manipulation functions; call
        motd_add() from the conf parser; call motd_clear() from the rehash
        routine; remove the no longer needed memory clearing and reloading
        stuff from the rehash service routine

        * ircd/motd.c: loads new API, including static internal functions
        to do allocation/deallocation, etc.

        * ircd/m_stats.c: use new motd_report() instead of
        report_motd_list()

        * ircd/m_motd.c: use new syntax for motd_send()

        * ircd/ircd.c: use new motd_init() function

        * ircd/Makefile.in (SRC): forgot to add motd.c to SRC in
        Makefile.(in); also ran make depend

        * include/motd.h: don't need config.h, but now do need time.h;
        define new structures and constants; redefine old API and define
        new functions

2000-11-22  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_user.c (register_user): use motd_signon() instead of
        calling m_motd; much cleaner this way

        * ircd/motd.c: write the new motd_* stuff to make MOTD handling
        less of a crock

        * ircd/m_motd.c: rewrite m{,s}_motd to call out to new motd_*
        functions

        * include/motd.h: define new MOTD API stuff

2000-11-20  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_reply.c (protocol_violation): rewrite
        protocol_violation so it'll actually work

        oh, yeah, use %s -> cptr->name, instead of %c -> cptr, so we get
        the client's real name in there.

        * ircd/m_motd.c (m_motd): Iso's addition of get_client_class(sptr)
        resulted in core dumps if NODEFAULTMOTD is defined, because m_motd
        gets called from register_user with a NULL sptr.  This is probably
        a design problem, but this bandaid will do for now...

2000-11-19  Isomer <isomer@coders.net>
        * ircd/ircd_reply.c: added 'protocol_violation', thus alerting us
        to problems in the server<->server protocol.

        * ircd/m_connect.c: allow remote connects with a port of '0'
        meaning to use the port in the config file.

        * ircd/m_create.c: Enable hacking protection, lets see how far we
        get.

        * ircd/m_error.c: The RFC says never accept ERROR from unreg'd
        clients, so we don't any more.

        * ircd/m_kill.c: The kill path is now made up of numnicks of servers,
        and the user@host is displayed of the victim.

        * ircd/m_map.c: reloaded 'dump_map'.

        * ircd/m_trace.c: allow per class T:

        * ircd/m_stats.c: allow local opers /remote stats anywhere on the 'net.

2000-11-17  Isomer <isomer@coders.net>

        * ircd/m_topic.c: Fixed bug where we'd only send to clients topics
        that were the *same* instead of different.  Oh the embarrasment!

        * ircd/IPcheck.c: Merged net's fix.

2000-11-02  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_whois.c: remove compiler warning by adding a newline to
        end of file

        * ircd/m_names.c: moved the flags up to s_user.h

        * ircd/m_join.c: call do_names instead of m_names; restructure
        ms_join to never transmute a JOIN into a CREATE, but use the TS in
        the JOIN (if present) to timestamp the channel

        * ircd/channel.c: send JOINs individually, instead of grouping
        them, so that we can send the channel's creation time

        * include/s_user.h: declare do_names()

2000-10-30  Isomer <isomer@coders.net>
        * ircd/m_oper.c: Fixed warning

2000-10-30  Isomer <isomer@coders.net>
        * ircd/m_oper.c: Fixed over agressive cut and no paste

2000-10-30  Isomer <isomer@coders.net>

        * ircd/m_topic.c: Restructured, fixed bug where topics on local
        channels are propergated (I forget who pointed this out to me, but
        thanks anyway).  Also to save bandwidth don't send the topic to
        users if the topic is already the same on the server (but still
        propergate to other servers).  X/W's "autotopic" feature must
        chew a lot of bandwidth, hopefully this will help reduce this.

        * doc/rfc1459.rfc: Updated documentation on /topic.

        * ircd/listener.c: snotice warnings about failed accept()'s
        potentially warning admins that they're running out of fd's.

        * ircd/stats.c, ircd/class.c: Removed /stats v, added number of
        people in a class in /stats y

        * ircd/m_create.c: Checks for timewarp hacking and squit's
        evil servers. (currently disabled)
        

2000-10-30  net <simms@lucida.qc.ca>
        
        * ircd/gline.c: Fixed various bugs Isomer left behind.

2000-10-26  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_join.c (m_join): reply on attempt to join a BADCHANed
        channel is now ERR_BANNEDFROMCHAN instead of ERR_BADCHANNAME

2000-10-24  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/channel.c: ok, now last mode rules; mode +ps will always
        result in +s (and won't send a mode if the channel is already +s);
        mode +sp will always result in +p; -n+n on a +n channel results in
        no mode change; -n+n on a -n channel results in a +n mode change;
        etc.

2000-10-23  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/channel.c: add "add" and "del" elements to ParseState to
        avoid not-too-pretty -p+s when +s is sufficient; fix a bug in
        mode_parse_limit that caused it to clear all channel modes
        prematurely; restructure mode_parse_mode to avoid calling
        modebuf_mode too early (ties in with first mentioned change);
        better logic for +p/+s mutual exclusivity; initialize "add" and
        "del" elements in mode_parse; send simple modes down to
        modebuf_mode after the loop in mode_parse

2000-09-28  Greg Sikorski <gte@atomicrevs.demon.co.uk>
        * ircd/m_names.c: Fixed a non-lethal logic error that 
        triggers an assert() in find_member_link while debugging.
        (Spotted by Maniac-).
2000-09-19  Thomas Helvey <helveytw@home.com>
        * ircd/s_conf.c: move K:lines to their own list and data
        structures, add supporting code.
        * ircd/m_stats.c: cleanup stats processing a bit move
        kline listing code to a new function, haven't figured
        out where it goes yet tho'
        * ircd/s_stats.c: added K:line bulk lister
        * include/s_conf.h: added new DenyConf struct
        * *[ch]: fixeup code that depended on changes

2000-09-17  Thomas Helvey <helveytw@home.com>
        * ircd/class.c: encapsulate class list
        * include/class.h: clean up classes
        * * fixup code that depended on changes

2000-09-17  Thomas Helvey <helveytw@home.com>
        * ircd/s_conf.c: add me to local conf
        * include/s_conf.h: move CONF_ME macro to chkconf.c
        * ircd/s_bsd.c: cleanup initialization, allow virtual host
        to be changed by rehash

2000-09-17  Thomas Helvey <helveytw@home.com>
        * include/class.h: add missing prototype
        * ircd/class.c: make argument to get_conf_class const

2000-09-17  Thomas Helvey <helveytw@home.com>
        * ircd/*.c: merged in changes from 2.10.10.pl12, cleanup
        merge conflicts.
        * ircd/*.h: merged in changes from 2.10.10.pl12, cleanup
        merge conflicts

2000-09-16  Thomas Helvey <helveytw@home.com>
        * ircd/s_conf.c: add code for server struct
        * ircd/client.c: copy of class.c sort of, new file for client
        specific operations, will move things here as appropriate,
        currently only one function is exported from here.
        * ircd/*.c: general logic cleanups, convert negatives to
        positives in places.

2000-09-16  Thomas Helvey <helveytw@home.com>
        * ircd/s_conf.c: add code for new crule data structs, strip quotes
        * ircd/crule.c: clean up scary casting a bit, type safety stuff
        * include/s_conf.h: add CRuleConf struct and support, remove
        unused constants
        * include/crule.h: type safety cleanups
        * ircd/*.c: fixup code that depended on stuff I changed

2000-09-15  Thomas Helvey <helveytw@home.com>
        * ircd/s_conf.c: start adding code for new conf data structs, changed
        listeners, admin line, motd lines, class lines. Move validate_hostent
        to resolver. General mayhem.
        * include/s_conf.h: new data structs and accessors
        * ircd/res.c: move validate_hostent here, rewrite, use regular
        expression for validation.
        * doc/example.conf: update docs for port

2000-09-14  Thomas Helvey <helveytw@home.com>
        * ircd/s_conf.c (conf_init): rewrite conf file parser, start to break
        up conf_init into managable chunks.
        * ircd/listener.c (set_listener_mask): fix logic bug core dump.
        * include/s_conf.h: add new data struct for local info (unwinding the mess).

2000-09-13  Thomas Helvey <helveytw@home.com>
        * ircd/list.c: put Clients in free lists, pre-allocate MAXCONNECTIONS
        local clients.
        * ircd/list.c: put SLinks in free lists
        * ircd/channel.c: put Memberships in free lists
        * ircd/ircd.c: rearrange initializations a bit in main
        Note: With these changes, ircd NEVER frees Clients, SLinks or
        Memberships. It will also rarely need to allocate new
        ones during net bursts and other disruptions. This should
        cut down on memory fragmentation a bit as well.

2000-08-30  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_names.c (do_names): pull-up from do_names fix in
        u2.10.10.pl11

2000-07-15  Perry Lorier       <Isomer@coders.net>
        * various: IP only k:'s and G:'s now do bit tests instead of two(!) 
                 match()'s.  Major Major cpu savings.  Also speed up the
                 other case slightly.  As a side effect you can now
                 k/Gline *@10.0.0.0/8.  I'll do bans tomorrow, it's nearing
                 3am.

2000-07-15  Perry Lorier       <Isomer@coders.net>
        * various: Fixed warnings after compiling on an alpha.
2000-07-09  Perry Lorier       <Isomer@coders.net>
        * doc/ircd.8: Applied grammitical changes by Liandrin, applied
                      changes suggested by various other people.
        * ircd/IPcheck.c: More bug fixes.  Current problem appears to be
                        that it gets a corrupt entry somehow.
2000-07-09  Greg Sikorski <gte@atomicrevs.demon.co.uk>
        * ircd/m_oper.c: Clean up compiler warning.

2000-07-08  Perry Lorier       <Isomer@coders.net>
        * doc/ircd.8: Updated the documentation, it was slightly out of date
                      being updated around 1989.
        * ircd/m_whois.c: Rewrote for clarity, and probably a bit more speed.
                          fixed a few minor glitches.
        * doc/rfc1459.unet: Updated.
        * ircd/IPcheck.c: Fixed more bugs.
        * ircd/s_bsd.c: We now keep track of servers we've conected.

2000-07-02  Perry Lorier       <Isomer@coders.net>
        * ircd/s_misc.c: Fixed remote IPcheck bug.  Ok, I'm a moron, so sue
                        me.  Thanks to Hektik, thanks thanks thanks thanks
                        thanks thanks thanks thanks thank thanks thank thanks

2000-07-01  Perry Lorier       <Isomer@coders.net>
        * ircd/s_conf.c: "Fixed" the "bug" where people would "evade" K:'s.
        * ircd/s_conf.c, include/IPcheck.h: Fixed compile warnings.

2000-06-22  Perry Lorier       <Isomer@coders.net>
        * ircd/IPcheck.c: Large chunks redone.
        * ircd/s_conf.c: Changes due to IPcheck - ONE nolonger supported,
                        single AND double digit limits are allowed now.
        * misc other: Changes to IPcheck.

2000-06-30  Perry Lorier       <Isomer@coders.net>
        * ircd/ircd.c: Fix command line parameter bugs.

2000-06-30  Perry Lorier       <Isomer@coders.net>
        * ircd/m_kill.c: Fixed bug with LOCAL_KILL_ONLY
        * ircd/m_nick.c: Tidied things up.

2000-06-12 Joseph Bongaarts <foxxe@trms.com>
        * ircd/m_stats.c: Iso forgot mo_stats when he added /stats v
        
2000-05-29  Perry Lorier       <Isomer@coders.net>
        * ircd/m_stats.c: add /stats v to do only the last part of the /trace
        * ircd/IPcheck.c: Cosmetic change, if we meddle with it enough do
                        you think it'll get bored and fix itself?

2000-06-09  Greg Sikorski <gte@atomicrevs.demon.co.uk>

        * ircd/m_names.c: Clean up compiler warnings.

2000-06-09  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/channel.c (mode_parse_client): don't send warning if
        there's not enough arguments for a +/-o/v; means the habit of
        doing "/mode #channel +oooooo bob" doesn't result in a bunch of
        error messages

2000-06-04  Greg Sikorski <gte@atomicrevs.demon.co.uk>

        * ircd/m_names.c: Re-factor code to remove unneccessary
        GlobalChannelList iteration every time someone joins a channel.

2000-06-02  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_user.c: add struct Gline * argument to register_user;
        look up global glines and repropagate them if necessary; send
        acknowledgement of gline to remote servers when registering users

        * ircd/s_serv.c (server_estab): don't send acknowledgement of
        local glines to remote servers; do send gline acknowledgement of
        bursted users

        * ircd/m_user.c (m_user): pass new struct Gline * argument to
        register_user

        * ircd/m_pong.c: pass new struct Gline * argument to register_user

        * ircd/m_nick.c (ms_nick): document protocol change

        * ircd/gline.c: support GLINE_LASTMOD

        * include/s_user.h: add struct Gline * argument to register_user

        * include/gline.h: add GLINE_LASTMOD to look up non-zero lastmods

        * ircd/s_conf.c (find_kill): add unsigned int argument to
        gline_lookup()

        * ircd/gline.c: add GLINE_GLOBAL to lookup or find only global
        glines; add unsigned int argument to gline_lookup()

        * include/gline.h: add GLINE_GLOBAL flag; add unsigned int
        argument to gline_lookup()

        * ircd/m_server.c: Resend jupe only when there is no %<lastmod>
        parameter, or when it falls out of bounds: see comments prior to
        call to jupe_resend(); call server_estab with struct Jupe
        parameter, so that we place the appropriate %<lastmod> in the
        appropriate place.

        * ircd/s_serv.c (server_estab): send %<lastmod> for introduced
        server, as well as for servers when we're sending the BURST

        * include/s_serv.h: add a struct Jupe * to the arguments for
        server_estab() so that we can send the appropriate lastmod
        parameter

        * ircd/m_gline.c (ms_gline): actually, this should be the
        slightest bit more efficient...

        * ircd/m_jupe.c (ms_jupe): actually, this should be the slightest
        bit more efficient...

        * ircd/m_gline.c (ms_gline): inhibit GLINE processing resends
        during netburst

        * ircd/m_jupe.c (ms_jupe): inhibit JUPE processing resends during
        netburst

        * ircd/channel.c (joinbuf_join): really remove user from local
        channels

2000-05-29  Perry Lorier       <Isomer@coders.net>
        * ircd/m_names.c: Removed redundant space. 
        * ircd/s_bsd.c: Fixed incorrect syntax on ERROR line.

2000-05-18  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_burst.c (ms_burst): er...that should have been a ",", not
        a " "

2000-05-04  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/channel.c: replace bogus assertions with returns, which is
        logically correct; only wipe out limit/key if they were originally
        set in the first place; remove user from channel when doing a
        PARTALL; only send MODE +o for user CREATEing channel if user is
        not MyUser--CREATE will only be used if the channel did not
        originally exist, therefore we can assume no one local is on the
        channel anyway, and we don't exactly need for the user to see an
        explicit +o for themselves

        * doc/readme.gline: describe the syntax of the GLINE command

        * doc/readme.jupe: update to reflect a couple of changes to JUPE

        * ircd/gline.c: don't propagate local changes

        * ircd/jupe.c: don't propagate local changes

        * ircd/m_gline.c (mo_gline): force local flag when deactivating
        glines with 0 lastmod

        * ircd/gline.c (gline_deactivate): G-lines with zero lastmod time
        are now removed instead of being deactivated

        * ircd/m_gline.c (ms_gline): make G-lines of the form "GLINE *
        -<mask>" be accepted

        * ircd/channel.c (send_channel_modes): deal with one of the last
        vestiges of sendbuf

        * ircd/m_burst.c (ms_burst): debugged ban processing; removed
        debugging hooks

        * ircd/channel.c (modebuf_extract): remove debugging
        sendto_opmask_butone calls

2000-05-03  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/channel.c: support a couple of new flags to support using
        mode_parse; fix some bugs with 0 struct ModeBuf *; implementation
        of modebuf_extract to extract added flags for use by ms_burst

        * include/channel.h: a couple of new flags to support using
        mode_parse inside ms_burst

        * ircd/m_burst.c (ms_burst): brand new implementation of BURST

        * ircd/m_endburst.c: add loop to processing of end_of_burst to
        free empty channels after the BURST is over.

        * ircd/m_server.c: convert to use new send.c functions--I wanted
        to rewrite it from scratch, but the logic's pretty complex; I may
        still rewrite it, though...

2000-05-02  Thomas Helvey <tomh@inxpress.net>

        * ircd/ircd.c: fix broken header include ordering

2000-05-02  Thomas Helvey <tomh@inxpress.net>
        
        * ircd/IPcheck.c: cleanups for ZenShadow's cleanups
        review emailed privately

        * include/IPcheck.h: removed unneeded include

2000-05-02  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_user.c (hunt_server): throw in a comment so I know what
        the sendto_one is for

        * include/querycmds.h (Count_unknownbecomesclient): convert to
        sendto_opmask_butone

        * ircd/send.c: start removing dead code

        * include/send.h: start removing dead code

        * ircd/m_rping.c: convert to sendcmdto_one / send_reply /
        hunt_server_cmd

        * ircd/m_rpong.c: convert to sendcmdto_one / send_reply

2000-05-01  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_stats.c: convert to sendcmdto_one / send_reply

        * ircd/m_kick.c: Completely reimplement m_kick

        * ircd/channel.c: send_user_joins removed; it was dead code,
        anyway...

2000-05-01  Perry Lorier <isomer@coders.net>
        * ircd/m_invite.c: Fix for the rest of m_invite.c, and again.
        * ircd/channels.c: My fix for the part problem.  Untested, probably
                won't work.  Can't be much worse than the current problem.
                it'll either work or core, take your pick.


2000-04-30  Perry Lorier <isomer@coders.net>
        * config/config-sh.in: Fix for CONNEXIT
        * ircd/s_{user,misc}.c: Fix for CONNEXIT
        * ircd/m_invite.c: Fix for incorrectly numnickified invite.
                        (Kev: Want to come talk to me about this?)

2000-04-30  Steven M. Doyle <steven@doyle.net>
        * ircd/ircd.c
          - general cleanups and readability enhancements
          - rewrite of setuid/chroot code.
          - server will no longer run as root
          - -DPROFIL compile option removed
          - Fixed IPcheck API calls
 
        * config/config-sh.in
          - Fixed up chroot compile options
          - Added options for debug and profile compiles
 
        * config/gen.ircd.Makefile
          - Support for new debug/profile options
 
        * ircd/Makefile.in
          - Support for new debug/profile options
 
        * ircd/ircd_signal.c
          - Removed -DPROFIL

        * include/IPcheck.h
          - Removed old API prototypes, added new ones
        
        * ircd/IPcheck.c
          - Readability cleanups (well, I -think-...)
          - Changed IPRegistryEntry.last_connect to a time_t.  The previously
            used unsigned short was probably causing interesting things after
            a client had been connected longer than about 65,535 seconds...
          - Removed old API functions.

        * ircd/whocmds.c
          - Removed IPcheck.h include
        
        * Additionally modified IPcheck API calls in:
          - ircd/m_nick.c
          - ircd/m_auth.c
          - ircd/s_bsd.c
          - ircd/s_conf.c
          - ircd/s_misc.c
          - ircd/s_serv.c
          - ircd/s_user.c
        
        
2000-04-30  Perry Lorier <isomer@coders.net>
        * ircd/s_bsd.c: Sigh. :)
        * ircd/m_mode.c: fix for modeless channels by poptix.

2000-04-29  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_join.c: reimplement JOIN in terms of struct JoinBuf

        * ircd/channel.c (clean_channelname): make clean_channelname also
        truncate long channel names

2000-04-28  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_create.c: reimplement CREATE in terms of struct JoinBuf

        * ircd/channel.c: implemented joinbuf_init, joinbuf_join,
        joinbuf_flush

        * include/channel.h: definitions and declarations for the struct
        JoinBuf abstraction

2000-04-29  Perry Lorier <isomer@coders.net>
        * ircd/s_bsd.c: Ok, so I thought I compiled and tested this...

2000-04-29  Perry Lorier <isomer@coders.net>
        * ircd/s_bsd.c: Add debugging code to IPcheck

2000-04-28  Kevin L. Mitchell  <klmitch@mit.edu>

        * include/ircd_reply.h (SND_EXPLICIT): use instead of RPL_EXPLICIT

        * ircd/ircd_reply.c (send_reply): use SND_EXPLICIT instead of
        RPL_EXPLICIT

        * ircd/m_userhost.c (m_userhost): add a dead code comment

        * ircd/m_desynch.c: forgot one...

        * ircd/m_rehash.c (mo_rehash): er, duplicates :)

        * ircd/m_proto.c (proto_send_supported): just change a comment so
        it doesn't show up in my scans

        * ircd/ircd_reply.c (send_reply): fix a slight bug...

        * ircd/s_numeric.c (do_numeric): use new sendcmdto_* functions,
        kinda hackish...

        * ircd/parse.c (parse_server): argument wrangling to make
        processing in do_numeric a little easier to deal with

        * ircd/s_serv.c (server_estab): SERVER should come from
        acptr->serv->up, not &me

        * ircd/m_lusers.c: accidentally left out sptr for a %C

        * ircd/send.c: hack to support doing wallchops...

        * ircd/m_whowas.c: convert to new send functions

        * ircd/m_whois.c: convert to new send functions

        * ircd/m_who.c: convert to new send functions

        * ircd/m_wallops.c: convert to new send functions

        * ircd/m_wallchops.c: convert to new send functions

        * ircd/m_version.c: convert to new send functions

        * ircd/m_userip.c: convert to new send functions

        * ircd/m_userhost.c: convert to new send functions

        * ircd/m_uping.c: convert to new send functions

        * ircd/m_trace.c: convert to new send functions

        * ircd/m_topic.c: convert to new send functions

        * ircd/m_time.c: convert to new send functions

        * ircd/m_squit.c: convert to new send functions

        * ircd/m_silence.c: convert to new send functions

        * ircd/m_settime.c: convert to new send functions

        * ircd/m_restart.c: convert to new send functions

        * ircd/m_rehash.c: convert to new send functions

        * ircd/m_privmsg.c: convert to new send functions

        * ircd/m_pong.c: convert to new send functions

        * ircd/m_ping.c: convert to new send functions

        * ircd/m_pass.c: convert to new send functions

        * ircd/m_opmode.c: convert to new send functions

        * ircd/m_oper.c: convert to new send functions

        * ircd/m_notice.c: convert to new send functions

        * ircd/m_nick.c: convert to new send functions

        * ircd/m_names.c: convert to new send functions

        * ircd/m_motd.c: convert to new send functions

        * ircd/m_mode.c: convert to new send functions

        * ircd/m_map.c: convert to new send functions

        * ircd/m_lusers.c: convert to new send functions

        * ircd/m_list.c: convert to new send functions

        * ircd/m_links.c: convert to new send functions

        * ircd/m_kill.c: convert to new send functions

        * ircd/m_jupe.c: convert to new send functions

        * ircd/m_invite.c: convert to new send functions

        * ircd/m_info.c: convert to new send functions

        * ircd/m_help.c: convert to new send functions

        * ircd/m_gline.c: convert to new send functions

        * ircd/m_error.c: convert to new send functions

        * ircd/m_endburst.c: convert to new send functions

        * ircd/m_die.c: convert to new send functions

        * ircd/m_destruct.c: convert to new send functions

        * ircd/m_defaults.c: convert to new send functions

        * ircd/m_connect.c: convert to new send functions

2000-04-28  Perry Lorier <isomer@coders.net>
        * RELEASE.NOTES: Describe a few more undocumented features.
        * config/config-sh.in: change the default paths for logging
        and the recommended number of channels.
        * include/supported.h: Rearrange slightly, added CHANTYPE's

2000-04-27  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_close.c: convert to send_reply

        * ircd/m_clearmode.c: convert to send_reply, sendcmdto_serv_butone

        * ircd/m_away.c: convert to send_reply and sendcmdto_serv_butone

        * ircd/m_admin.c: convert to send_reply and hunt_server_cmd

        * ircd/s_user.c (hunt_server_cmd): new hunt_server replacement
        that takes cmd and tok arguments, etc.  NOTE: THIS IMPLEMENTATION
        HAS A MAJOR HACK!!!  The whole hunt_server architecture should be
        carefully rethought...

        * ircd/s_stats.c (hunt_stats): use new hunt_server_cmd

        * include/s_user.h: hunt_server_cmd -- replacement for hunt_server

        * ircd/s_misc.c: *sigh* 2.10.10 doesn't support squitting by
        numeric nick; therefore, we have to use the server name

        * ircd/m_squit.c (ms_squit): allow to squit by server numeric nick

        * ircd/send.c: fix minor bugs

        * ircd/s_user.c (check_target_limit): mark dead code so I filter
        it when I grep

        * ircd/s_serv.c (exit_new_server): mark dead code so I filter it
        when I grep

        * ircd/parse.c: mark dead code so I filter it when I grep

        * ircd/map.c: mark dead code so I filter it when I grep

        * ircd/ircd.c: mark dead code so I filter it when I grep

        * ircd/ircd_relay.c: convert over to new sendcmdto_*, send_reply
        functions

        * ircd/channel.c: mark dead code so I filter it when I grep

        * ircd/s_stats.c: use send_reply instead of sendto_one w/rpl_str;
        hope I'm not stepping on toes...

        * ircd/s_conf.c: more sendto_opmask_butone / send_reply
        conversions; use ircd_snprintf in a couple of cases to negate the
        possibility of buffer overflow

2000-04-26  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/channel.c: convert as much as possible to new send
        semantics

        * ircd/send.c (sendcmdto_common_channels): fix a subtle bug --
        test member->user->from->fd, not from->fd

        * ircd/gline.c (gline_add): go ahead and add badchans; we just
        won't look for them in m_gline; this way, they always work...

        * ircd/jupe.c: use ircd_vsnprintf conversion specifiers

        * ircd/gline.c: since write_log now uses ircd_vsnprintf, use
        ircd_vsnprintf conversion specifiers

        * ircd/support.c (write_log): use ircd_vsnprintf for write_log, so
        I have my conversion specifiers

        * ircd/gline.c (do_gline): use send_reply for ERR_YOUREBANNEDCREEP

        * ircd/send.c (sendcmdto_flag_butone): explicitly send WALLOPS to
        local users

        * ircd/s_serv.c (exit_new_server): rewrite exit_new_server to be a
        little less brain-dead

        * ircd/s_misc.c: use sendcmdto_one, sendrawto_one, and send_reply

        * ircd/s_debug.c: use send_reply with RPL_EXPLICIT for
        RPL_STATSDEBUG

        * ircd/res.c (cres_mem): use send_reply with RPL_EXPLICIT for
        RPL_STATSDEBUG

        * ircd/list.c (send_listinfo): use send_reply with RPL_EXPLICIT
        for RPL_STATSDEBUG

        * ircd/m_pong.c: use RPL_EXPLICIT for ERR_BADPING

        * ircd/ircd.c: use RPL_EXPLICIT for ERR_BADPING

        * ircd/s_user.c (register_user): use RPL_EXPLICIT for
        ERR_INVALIDUSERNAME

        * ircd/ircd_reply.c (send_reply): support RPL_EXPLICIT

        * include/ircd_reply.h (RPL_EXPLICIT): somewhat of a hack to mark
        a numeric as needing to use an explicit pattern, which will be the
        first argument in the variable argument list

        * ircd/s_user.c: use sendrawto_one instead of sendto_one to send
        non-prefixed nospoof PING

        * ircd/s_bsd.c: use sendrawto_one instead of sendto_one to send
        non-prefixed SERVER login

        * ircd/ircd.c (check_pings): fix last sendto_one calls (except for
        a numeric usage further up)

        * include/send.h: declare sendrawto_one

        * ircd/send.c (sendrawto_one): new function to use ONLY for
        non-prefixed commands, like PING to client, or PASS/SERVER on
        server registration

2000-04-25  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/ircd_snprintf.c (doprintf): implement %H for possible
        future expansion (channel numerics?)

        * include/ircd_snprintf.h: added documentation to # to explain use
        with %C; added documentation for : to explain use with %C; added
        documentation for %H for channels

        * ircd/whocmds.c: use send_reply

        * ircd/userload.c: use sendcmdto_one

        * ircd/uping.c: use sendcmdto_one

        * ircd/send.c: use new flags to %C format string; ':' prefixes
        client name with a colon for local connects, '#' uses
        nick!user@host form for local connects

        * ircd/s_user.c: use send_reply, sendto_opmask_butone,
        sendcmdto_one, sendcmdto_serv_butone, sendcmdto_flag_butone

        * ircd/s_serv.c: use sendcmdto_one, sendto_opmask_butone

        * ircd/s_bsd.c: use sendto_opmask_butone, send_reply,
        sendcmdto_one

        * ircd/s_auth.c: use sendto_opmask_butone

        * ircd/res.c: use sendcmdto_one

        * ircd/ircd_snprintf.c (doprintf): minor bug fixes and some
        debugging assertions

2000-04-24  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/support.c: dumpcore is no longer used, so get rid of it

        * ircd/parse.c: use send_reply, sendcmdto_one

        * ircd/map.c: use send_reply

        * ircd/listener.c: use send_reply

        * ircd/jupe.c: use sendto_opmask_butone, send_reply

        * ircd/ircd_reply.c: use send_reply

        * ircd/ircd.c: use sendto_opmask_butone

        * ircd/gline.c: use sendto_opmask_butone, send_reply

        * ircd/ircd_snprintf.c (doprintf): make it deal with incompletely
        registered clients; make FLAG_ALT print nick!user@host; make
        FLAG_COLON print :blah

        * ircd/class.c (report_classes): use send_reply instead of
        sendto_one

        * ircd/hash.c (m_hash): replace sendto_one with sendcmdto_one

        * ircd/IPcheck.c (ip_registry_connect_succeeded): replace
        sendto_one with sendcmdto_one

2000-04-21  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/send.c: clean up logic in sendcmdto_channel_butone; use
        MyConnect() instead of IsServer() in sendcmdto_flag_butone; define
        sendcmdto_match_butone

        * include/send.h: declare sendcmdto_match_butone

2000-04-20  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/jupe.c: update to use send_reply()

        * ircd/gline.c: update to use send_reply()

        * include/ircd_reply.h: declare send_reply

        * ircd/ircd_reply.c (send_reply): send_error_to_client, but for
        replies; uses ircd_snprintf

        * ircd/send.c: added comments to redirect searchers to appropriate
        sendcmdto_* function; moved new functions to end of file; added
        explanatory comments; reordered arguments; defined new functions
        mentioned below

        * ircd/m_jupe.c: reorder arguments to sendcmdto_* functions

        * ircd/m_gline.c: reorder arguments to sendcmdto_* functions

        * ircd/jupe.c: reorder arguments to sendcmdto_* functions

        * ircd/gline.c: reorder arguments to sendcmdto_* functions

        * include/send.h: reorder arguments, add explanatory comments,
        declare new functions sendcmdto_flag_butone, sendto_opmask_butone,
        and vsendto_opmask_butone

2000-04-19  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/send.c: define sendcmdto_channel_butone, wrote a simplified
        vsendto_op_mask that uses '*' instead of the receiving client
        nickname

        * include/send.h: declare sendcmdto_channel_butone; takes a skip
        argument that allows you to skip (or not to skip) deaf users,
        users behind bursting servers, and non channel operators

2000-04-17  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/send.c: new sendcmdto_channel_butserv -- note that old
        sendto_channel_butserv has a subtle bug; also wrote
        sendcmdto_common_channels.

        * include/send.h: declare new sendcmdto_* functions

        * ircd/jupe.c: support local deactivations of jupes

        * ircd/gline.c: support local deactivations of glines

        * include/jupe.h: JUPE_LDEACT allows jupes to be locally
        deactivated; if they aren't locally deactivated, then it slaves to
        the net-wide activation status; JupeIsRemActive() tests only
        whether the jupe is active everywhere else

        * include/gline.h: GLINE_LDEACT allows glines to be locally
        deactivated; if they aren't locally deactivated, then it slaves to
        the net-wide activation status; GlineIsRemActive() tests only
        whether the gline is active everywhere else

        * ircd/gline.c: detect overlapping G-lines; if an existing, wider
        gline expires after the new one will, we drop the new one,
        otherwise we add the G-line after that one (so the wide one will
        apply first); if the new one contains an existing G-line and if it
        will expire after the existing one, we drop the existing one to
        save memory

        * ircd/m_gline.c (mo_gline): opers could issue remote local
        glines when CONFIG_OPERCMDS was off; fixed

2000-04-16  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_jupe.c (mo_jupe): allow target argument to be dropped if
        this is a local JUPE

        * ircd/gline.c: add flags argument to gline_activate and
        gline_deactivate for future expansion

        * ircd/m_gline.c: pass flags to gline_activate and
        gline_deactivate

        * include/gline.h: add flags argument to gline_activate and
        gline_deactivate

        * ircd/jupe.c: add flags argument to jupe_activate and
        jupe_deactivate for future expansion

        * include/jupe.h: add flags argument to jupe_activate and
        jupe_deactivate

        * ircd/m_jupe.c: pass a flags argument to jupe_add instead of
        local, active; pass flags to jupe_activate and jupe_deactivate

        * include/gline.h: remove dead code

        * ircd/gline.c: make gline expire times relative to CurrentTime,
        since that should be monotonically increasing, instead of
        TStime(), which can be set backwards, and which can therefore
        cause an expire time to increase; make local glines be removed
        instead of just deactivated; don't let gline_find() look for
        user@host glines if the mask being looked up is a channel mask

        * ircd/send.c (vsendcmdto_one): forgot to account for the case
        where origin is a server and destination is a user

        * ircd/jupe.c: make jupe expire times relative to CurrentTime,
        since that should be monotonically increasing, instead of
        TStime(), which can be set backwards, and which can therefore
        cause an expire time to increase; make local jupes be removed
        instead of just deactivated

        * ircd/ircd_snprintf.c: d'oh, thanks for catching that; short for
        limit is fine.  any other warnings I should know about?

2000-04-15  Thomas Helvey <tomh@inxpress.net>

        * ircd/*.c: const correctness and type safety cleanups to
        get code to compile with C++ compiler. Still has
        signed/unsigned comparison warnings.

2000-04-15  Greg Sikorski <gte@atomicrevs.demon.co.uk>

        * ircd/userload.c: change <sys/time.h> include to <time.h> for
          portability.

2000-04-14  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_gline.c (mo_gline): d'oh, target isn't a numeric; use %C
        and convert acptr...

        * ircd/s_user.c: move gline_lookup function call into
        register_user, where it'll have a username to lookup!

        * ircd/m_gline.c: modify to utilize new sendcmdto_* series of
        functions; also stuff send_error_to_client into return clauses

        * ircd/m_jupe.c: modify to utilize new sendcmdto_* series of
        functions; also use send_error_to_client where that makes sense

        * ircd/jupe.c: modify to utilize new sendcmdto_* series of
        functions; also use send_error_to_client where that makes sense

        * ircd/gline.c: modify to utilize new sendcmdto_* series of
        functions; also fix gline_lookup() to deal properly with remote
        clients--boy, do struct Client and struct User need to be cleaned
        up!

        * ircd/ircd_snprintf.c (doprintf): a dest of &me is a server,
        too...

        * ircd/send.c: wrote sendcmdto_one(), vsendcmdto_one(), and
        sendcmdto_serv_butone(), all utilizing the %v conversion of
        ircd_snprintf()

        * include/send.h: define IRC_BUFSIZE, max size of a message;
        declare sendcmdto_one(), vsendcmdto_one(), and
        sendcmdto_serv_butone()

        * include/msg.h: define all the CMD_* constants needed to utilize
        the new sendcmdto_* series of functions

        * ircd/Makefile.in (SRC): list ircd_snprintf.c; run make depend

        * ircd/gline.c: remove old, dead code.

        * ircd/m_gline.c (mo_gline): disallow setting of global G-lines
        unless CONFIG_OPERCMDS is enabled; disallow listing of all G-lines
        (don't advertise proxies); remove dead code

        * ircd/parse.c: oper handler for JUPE only lists jupes unless
        CONFIG_OPERCMDS is enabled

        * ircd/m_jupe.c (mo_jupe): don't compile mo_jupe() if
        CONFIG_OPERCMDS is not enabled; we'll disable it in parse.c

        * ircd/m_opmode.c (mo_opmode): if CONFIG_OPERCMDS is not enabled,
        always return ERR_DISABLED

        * ircd/m_clearmode.c (mo_clearmode): if CONFIG_OPERCMDS is not
        enabled, always return ERR_DISABLED

        * ircd/s_err.c: add error message to indicate disabled commands

        * include/numeric.h (ERR_DISABLED): to indicate disabled commands

        * doc/Configure.help: add documentation for CONFIG_OPERCMDS

        * config/config-sh.in: add CONFIG_OPERCMDS, default both it and
        CONFIG_NEW_MODE to 'y' for now

        * ircd/gline.c (gline_list): fix a minor formatting bogon

        * BUGS: since I fixed that bug, might as well mark it fixed.

        * ircd/m_join.c: look up badchans with GLINE_EXACT

        * ircd/m_gline.c: fix parc count problems; look up existing
        G-lines with GLINE_EXACT; only set new lastmod when
        activating/deactivating existing glines if old lastmod was not 0

        * ircd/gline.c: forgot to copy the gline reason over; don't
        propagate a gline with 0 lastmod if origin is user; add
        GLINE_EXACT to force exact matching of gline mask

        * ircd/ircd_snprintf.c (doprintf): forgot to deal with the zero
        flag properly

        * ircd/s_conf.c (find_kill): gline_find() takes a char *userhost,
        but gline_lookup() actually takes a client--d'oh.

2000-04-13  Thomas Helvey <tomh@inxpress.net>
        * ircd/IPcheck.c: Back port BLMet's bugfix from 2.10.10

2000-04-13  Greg Sikorski <gte@atomicrevs.demon.co.uk>

        * ircd/whocmds.c: Don't make idle flag default in /who, to prevent:
          "/who * x"
          "Gte3 H*iwg Gte@212.49.240.217 :1 :0 I am the one that was."
          (Found by Plexus).

        * ircd/whocmds.c: Change idle time calc from socket idle to user
          idle.

2000-04-13  Kevin L. Mitchell  <klmitch@mit.edu>

        * config/aclocal.m4 (unet_CHECK_TYPE_SIZES): check size of void *,
        too, for ircd_snprintf.c

        * include/ircd_snprintf.h: documentation for ircd_(v)snprintf, in
        comments; mostly descended from the Linux manpage for printf, but
        also documenting the extensions.

        * ircd/ircd_snprintf.c: NULL dest is equivalent to going to a
        client; make 'q' be the same as 'L'; remove __inline__; only
        define EXTENSION if HAVE_LONG_LONG is defined

        * include/handlers.h: declare m_gline()

        * ircd/parse.c: gline can be called by users, but it only lists
        the glines.

        * ircd/s_user.c (set_nick_name): resend gline if a remote server
        introduces a glined client

        * ircd/s_serv.c (server_estab): burst glines, too

        * ircd/gline.c: fix up all the expire times to be offsets;
        simplify gline_resend()

        * ircd/m_gline.c: begin coding replacements for ms_gline(),
        mo_gline(), and m_gline()

        * ircd/gline.c (gline_add): allow *@#channel to work correctly;
        also, prohibit local BADCHANs if LOCAL_BADCHAN not defined

2000-04-13  Greg Sikorski <gte@atomicrevs.demon.co.uk>

        * tools/Bouncer/*: Add comments/documentation/tags.
        * tools/Bouncer/*: Add debug defines, make task fork().

2000-04-12  Thomas Helvey <tomh@inxpress.net>
        * ircd/s_err.c: Cleanup s_err.c make one table so we
        don't have to do anything tricky to get an error string.

2000-04-12  Greg Sikorski <gte@atomicrevs.demon.co.uk>
        * Add port bouncer for http (x/w)

2000-04-12  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/s_conf.c (find_kill): replaced call to find_gline() with a
        call to gline_find(); also used GlineReason() instead of direct
        reference to structure member

        * ircd/m_join.c (m_join): replace bad_channel() calls with calls
        to gline_find(name, GLINE_BADCHAN), and also check to see if gline
        is active

        * ircd/channel.c: nothing seems to be called anywhere...

        * ircd/s_err.c: update a couple of replies to dovetail with new
        semantics

        * ircd/gline.c: begin complete re-implementation of gline.c along
        the lines of the final design of jupe.c

        * include/gline.h: begin complete re-implementation of gline.c
        along the lines of the final design of jupe.c

        * ircd/channel.c (mode_process_clients): fix "Deop of +k user on
        %s by %s" message...

        * ircd/ircd_snprintf.c: my new snprintf()-like functions

        * include/ircd_snprintf.h: my new snprintf()-like functions

2000-04-11  Thomas Helvey <tomh@inxpress.net>
        * ircd/IPcheck.c: removed old dead code
        * ircd/s_user.c (send_user_info): removed non-standard
          user not found message for userhost/userip

2000-04-11  Greg Sikorski <gte@atomicrevs.demon.co.uk>

        * ircd/s_err.c: Added missing quotes to ERR_DONTCHEAT numeric.
        * doc/p10.html: Work on chapter 4.

2000-04-10  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/channel.c (mode_parse_client): fix coredump on /mode
        #foobar +o nosuchnick

2000-04-10  Perry Lorier  <Isomer@coders.net>
        * BUGS: Added bug.

2000-04-09  Thomas Helvey <tomh@inxpress.net>
        * include/IPcheck.h: fix prototype
        * ircd/s_user.c: fix usage of IPcheck_remote_connect
        * ircd/IPcheck.c: removed unused args

2000-04-09  Thomas Helvey <tomh@inxpress.net>
        * include/IPcheck.h: add proto for IPcheck_expire

        * ircd/IPcheck.c: Rewrote

        * ircd/ircd.c: Add IPcheck_expire to main message loop

        * ircd/s_user.c: Redo target hashing, refactor target code

        * include/numeric.h: Cleaned up numerics, added which ones are
        in use by other networks and what they are in use for.

        * ircd/channel.c: cleaned can_join(), allow anyone through anything
        if /invited, simplified the function.  Opers overusing OPEROVERRIDE
        will get a message explaining to them not to cheat.

        * ircd/m_join.c: cleaned up the various join functions, should be
        a lot more efficient.  Still needs work.  Now assumes that s<->s
        won't send it a JOIN 0.  Service coders - note this and tread with
        care.

        * ircd/m_stats.c: added Gte-'s stats doc patch.

        * ircd/m_version.c: /version now returns the 005 numeric as well.
        as requested by Liandrin.


2000-04-07  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_clearmode.c: add include for support.h for write_log()

        * configure: move ircd/crypt/* to tools/*

2000-04-06  Thomas Helvey <tomh@inxpress.net>
        * ircd/s_auth.c: Shorten auth connect timeout to 60 seconds
          set client host to server alias if connection from localhost

2000-04-06  Perry Lorier <isomer@coders.net>
        * ircd/ircd.c: Fix core during pinging (oops)
        
2000-04-06  Perry Lorier <isomer@coders.net>
        * ircd/send.c: fixed wrong ident being sent to channels bug.
        * include/numerics.h: Updated some of the numerics from other
        networks.  Flagged some as 'unused' by undernet.

2000-03-30  Perry Lorier <isomer@coders.net>
        * ircd/ircd.c: Lets see if this helps the ping problem at all.
        * ircd/whocmds.c, /doc/readme.who: Added %l specifier to get idle
        time for local clients. (as requested), extended who now returns all
        the flags (@+!) so you can tell the complete state of a client.

2000-03-30  Thomas Helvey <tomh@inxpress.net>
        * m_rping.c m_rpong.c: add Gte's rping/rpong fixes

2000-03-30  Perry Lorier <isomer@coders.net>
        * ircd/parse.c: oops, missed opers.

2000-03-30  Perry Lorier <isomer@coders.net>
        * ircd/parse.c: fixed mystifying ping bug thats been plaguing us
        for so long.  Remember: m_ping MUST be in the parse array. :)

2000-03-30  Perry Lorier <isomer@coders.net>
        * ircd/ircd.c: test in check_pings was wrong.  I move that we
        disallow cvs commit after 10pm localtime....

2000-03-30  Perry Lorier <isomer@coders.net>
        * ircd/m_pong.c: Fix it for servers too.

2000-03-30  Perry Lorier <isomer@coders.net>
        * ircd/m_pong.c: Fix ping timeout bugs

2000-03-30  Perry Lorier <isomer@coders.net>
        * ircd/channel.c: Bans had CurrentTime in their when field instead
        of TStime()

2000-03-31  Thomas Helvey <tomh@ixpress.net>
        * ircd/numnicks.c (SetXYYCapacity): fix for extended
        numerics.

2000-03-30  Perry Lorier <isomer@coders.net>
        * ircd/m_nick.c: send kills both ways so when we add nick change
        on collision we don't desync the network.

        * ircd/map.c: Fixup the map a bit more.

2000-03-31  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_clearmode.c (do_clearmode): Log the CLEARMODE to OPATH

        * ircd/m_opmode.c: Log the mode changes to OPATH

        * ircd/channel.c (modebuf_flush_int): Log the mode changes to
        OPATH

        * include/channel.h (MODEBUF_DEST_LOG): Log the mode changes to
        OPATH

        * doc/Configure.help: help text for CONFIG_LOG_OPMODE / OPATH

        * config/config-sh.in: added OPATH for opmode log file

        * ircd/m_clearmode.c (do_clearmode): updated uses of
        modebuf_mode_string() for the new usage

        * ircd/channel.c: added flag MODE_FREE and an int argument to
        modebuf_mode_string() to indicate that the string must be free'd;
        updated calls to modebuf_mode_string() for the new usage; called
        collapse(pretty_mask()) on the ban string and use allocated memory
        for it; added ban list length accounting; fixed a number of small
        bugs in ban processing

        * include/channel.h: added flag MODE_FREE and an int argument to
        modebuf_mode_string() to indicate that the string must be free'd

        * ircd/m_clearmode.c (do_clearmode): made sure clearmode removed
        keys and limits that are set

2000-03-30  Perry Lorier <isomer@coders.net>
        * ircd/ircd.c: rewrote check_pings() for maintainability
        and speed.  Also changed quit msg's so they don't have
        redundant nick[host] info in them.

        * ircd/send.c: Changed write errors to report what error
        occured (if possible).

        * ircd/gline.c: added gline comment to the quit.

        * ircd/m_server.c: Added suggestions to server quits mentioning
        what went wrong so the admin can fix it earlier instead of asking
        questions...

        * ircd/map.c: Changed m_map() to hide numerics, show a * beside
        servers that aren't fully burst yet.  And show '(--s)' for servers
        where its not sure.

        * doc/example.conf: Fixed wrapped U:

2000-03-30  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/m_mode.c (ms_mode): implemented a new m_mode in terms of
        mode_parse() (version selectable at compile time)

        * ircd/m_clearmode.c (mo_clearmode): clean_channelname(parv[1])

        * ircd/m_opmode.c (mo_opmode): clean_channelname(parv[1])

        * config/config-sh.in: add new config option to enable new m_mode
        implementation

        * doc/Configure.help: add documentation for new config option
        CONFIG_NEW_MODE

        * ircd/channel.c (mode_parse_client): /opmode #foobar -o -- 461
        MODE -v : Not enough parameters

        * ircd/m_clearmode.c (do_clearmode): do_clearmode() would remove
        +k and +l even if they weren't set...

        * ircd/m_opmode.c: implement the OPMODE command using mode_parse()

        * ircd/channel.c: make mode_process_clients() clear the DEOPPED
        flag; fix +s+p exclusivity; add MODE_ADD/MODE_DEL to flag list
        for; test the 0-th member, not the i-th member, of the client
        change state stuff

        * ircd/m_clearmode.c (do_clearmode): use the new
        mode_invite_clear() function

        * ircd/channel.c: cleared up all the compile-time warnings and
        errors

        * include/channel.h: added declarations for mode_ban_invalidate()
        and mode_invite_clear()

        * ircd/channel.c: finished mode_parse(), then broke it up into a
        dozen or so helper functions to make the code easier to read

2000-03-29  Thomas Helvey <tomh@inxpress.net>
        * ircd/ircd.c: refactor server initialization a bit, use
        getopt for parsing command line, refactor init_sys, main,
        and other bits.

        * ircd/s_bsd.c: add functions for initialization to clean
        up logic a bit and remove duplicated code.

        * include/ircd.h: add struct for server process related
        variables.

2000-03-29  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/channel.c: initial definition of mode_parse(); flags to
        prevent doing the same thing multiple times; helper method
        send_notoper() to send a "Not oper"/"Not on channel" notice

        * include/channel.h: declare mode_parse() and helper flags

        * ircd/channel.c (modebuf_flush_int): fiddled with timestamp
        sending to match the current action of set_mode() closely enough
        that hopefully there won't be major conflicts

        * ircd/channel.c (modebuf_flush_int): consolidated the mode string
        building logic, reversed the order of the arguments to mode
        commands to have '-' preceed '+'

2000-03-29  Thomas Helvey <tomh@inxpress.net>
        * ircd/s_bsd.c (add_connection): don't disable socket options
        let OS tune itself and allow important performance tweaks to 
        work.

2000-03-28  Kevin L. Mitchell  <klmitch@mit.edu>

        * ircd/channel.c (modebuf_flush_int): use %d, not %-15d; I got
        confused by set_mode, which is doing some really weird logic;
        guess what I'm going to rewrite next?  ;)

2000-03-28  Kevin L. Mitchell  <klmitch@emc.com>

        * include/channel.h: added MODE_SAVE for the bounds checking stuff
        in modebuf_flush

        * ircd/channel.c: make modebuf_flush into modebuf_flush_int and
        make it do bounds checking on the buffer; all modes are sent only
        if the all parameter is 1; modebuf_flush is the exported wrapper

        * include/channel.h: add BOUNCE, renumber flags to get a little
        more space

        * ircd/channel.c (modebuf_flush): don't overload HACK2, add
        BOUNCE; send DESYNCH message

2000-03-27  Kevin L. Mitchell  <klmitch@emc.com>

        * ircd/m_clearmode.c (do_clearmode): only mark the modes the
        channel actually has in effect for deletion

        * ircd/channel.c: added explanatory comments to all added
        functions; made flushing take place at the correct place even if
        the MODEBUF_DEST_DEOP flag is set; rewrote build_string() helper
        to bash some stupid bugs; made modebuf_flush() return if ModeBuf
        is empty, fixed the apparent source, removed some bogus string
        termination code, properly terminate the mode strings, add support
        for HACK2 and HACK3, made limit strings not be sent if the limit
        is being removed, changed where '+' and '-' come from in sent
        strings, added support for DEOP flag, set up bouncing code for
        HACK2

        * ircd/Makefile.in: ran make depend

        * include/channel.h: added new defines for future functionality,
        made modebuf_flush() return int so I can use tail recursion

        * ircd/m_clearmode.c: add msg.h to includes; other misc cleanups
        to make it all compile

        * ircd/m_opmode.c: add msg.h to includes...

        * ircd/m_clearmode.c: implemented mo_clearchan()/ms_clearchan()

        * ircd/channel.c (modebuf_flush): realized I forgot to
        nul-terminate addbuf/rembuf properly...

        * ircd/m_clearmode.c (do_clearmode): wrote do_clearmode()...

        * ircd/channel.c (modebuf_flush): correct sendto_server_butone to
        sendto_serv_butone--blah^2

        * ircd/send.c (sendto_serv_butone): stupid comments confused me

        * ircd/channel.c (modebuf_flush): if there are no mode changes to
        propagate, we're done...

        * ircd/channel.c (modebuf_flush): duh; it's sendto_server_butone,
        not sendto_all_butone

        * ircd/m_clearmode.c: define skeleton for m{o,s}_clearmode

        * ircd/m_opmode.c: define skeleton for m{o,s}_opmode

        * ircd/Makefile.in (SRC): added m_opmode() and m_clearmode() to
        the list

        * ircd/parse.c: added messages for opmode and clearmode

        * include/handlers.h: added declarations for mo_opmode(),
        ms_opmode(), mo_clearmode(), and ms_clearmode()

        * include/msg.h: define MSG_OPMODE, TOK_OPMODE, MSG_CLEARMODE, and
        TOK_CLEARMODE

        * include/channel.h (MODEBUF_DEST_OPMODE): Define the
        MODEBUF_DEST_OPMODE flag

        * ircd/channel.c (modebuf_flush): added new flag,
        MODEBUF_DEST_OPMODE; causes channel MODE/HACK(4) notice to appear
        to originate from source's server (or source itself, if
        IsServer(source)); also causes a server-level MODE to be sent as
        OPMODE instead

        * include/channel.h: defined MODEBUF_DEST_SERVER,
        MODEBUF_DEST_HACK4

        * ircd/channel.c: Add another argument to build_string() to handle
        numeric nicks; implemented MODEBUF_DEST_SERVER to send MODEs to
        servers; implemented MODEBUF_DEST_HACK4 to cause HACK(4) notices
        to be sent out

2000-03-27  Perry Lorier <isomer@coders.net>

        * ircd/s_bsd.c: fixed missing 'u' typo.

2000-03-26  Kevin L. Mitchell  <klmitch@emc.com>

        * ircd/channel.c: implement modebuf_init(), _mode(), _mode_uint(),
        _mode_string(), _mode_client(), _flush(); also implemented a
        simple build_string()

        * include/channel.h: added definition of ModeBuf, modebuf_*
        manipulation functions, and a couple of helper macros