aboutsummaryrefslogtreecommitdiff
path: root/src/util/network.c
Commit message (Collapse)AuthorAge
* -fix (C) noticesChristian Grothoff2016-01-19
|
* automatically clean up left-over Unix domain socket files when trying to ↵Christian Grothoff2015-07-13
| | | | bind (fixes #3723)
* fix #3869: outdated FSF addressChristian Grothoff2015-06-30
|
* Add missing const to the argument of the readiness checkersLRN2015-04-09
|
* Wait on actual HANDLE, not the structure that contains itLRN2015-04-09
|
* -trying to fix warningsChristian Grothoff2015-03-23
|
* generally use GNUNET_assert() instead of GNUNET_abort() to also log the errorChristian Grothoff2015-02-28
|
* -signal connection failure to receive even if receive is triggered after ↵Christian Grothoff2015-02-21
| | | | failure is observed
* be more careful with errno, so we can be sure it is not overwritten by ↵Christian Grothoff2015-02-21
| | | | 'free()' and the like
* -bringing copyright tags up to FSF standardChristian Grothoff2015-02-07
|
* -make sure we get the correct errno back after GNUNET_NETWORK_socket_create()Christian Grothoff2015-01-25
|
* W32: Remove copy-pasted stray incrementLRN2015-01-24
|
* Add GNUNET_NETWORK_fdset_handle_set_firstLRN2015-01-24
|
* Add support for selecting on W32 eventsLRN2015-01-24
|
* use enough space for storing full addressChristian Grothoff2015-01-17
|
* Grothoff's W32 select slist elimination and prettificationLRN2014-12-14
|
* Grothoff's massive indentation and documentation fixes in util/network.cLRN2014-12-14
|
* adding TCP_STEALTH support to TCP pluginChristian Grothoff2014-11-02
|
* - Do not try to delete empty string directories.Sree Harsha Totakura2014-02-14
|
* Create UNIX domain sockets as abstract sockets when running in LINUX and theSree Harsha Totakura2014-02-09
| | | | | option USE_ABSTRACT_SOCKETS is present in configuration.
* -clean up network select code, avoid insane #ifdefing for MINGWChristian Grothoff2013-12-13
|
* -fix off-by-one so we can also select on FD 0 (stdin)Christian Grothoff2013-12-13
|
* -fix #3211Christian Grothoff2013-12-13
|
* -fix warningChristian Grothoff2013-10-09
|
* removing remenants of abstract unix domain socket handling, this finishes ↵Christian Grothoff2013-10-09
| | | | addressing #2887
* changing UNIX domain socket access control to file permissions checks, ↵Christian Grothoff2013-10-09
| | | | instead of UDS credentials (#2887)
* -switching (again) to named sockets, see #2887Christian Grothoff2013-10-09
|
* -remove trailing whitespaceChristian Grothoff2013-10-06
|
* *sigh* ... Matthias Wachs2013-10-01
|
* -towards addressing #3047, note this causes the code to FTBFSChristian Grothoff2013-09-30
|
* -undo #29640, somehow causes problemsChristian Grothoff2013-09-27
|
* moving from abstract unix domain socket paths to normal unix domain socket ↵Christian Grothoff2013-09-26
| | | | paths (#2887), removing now unnecessary/dead flags argument to GNUNET_NETWORK_socket_bind
* Applying 1st patch from Andrew Cann (see also #2887).Christian Grothoff2013-09-26
| | | | | | | | | | | | https://canndrew.org/misc/trim_abstract_socket_paths.diff [^] This changes the addrlen argument passed to the connect, bind and sendto syscalls to reflect the length of the string in sockaddr_un.sun_path. I was trying to talk to GNUnet services using socat and discovered that it expects abstract sockets to be created this way and won't connect otherwise. Looking at some other programs that I could talk to using socat (dbus, X11), this is how they create abstract sockets as well.
* -check return values, doxygen, code cleanupChristian Grothoff2013-09-26
|
* -fix shortening routine to deal with longer hashesChristian Grothoff2013-09-20
|
* add GNUNET_NETWORK_socket_free_memory_only_ API to allow getting raw FD and ↵Christian Grothoff2013-09-16
| | | | releasing struct without leaking 'addr' member
* Fix timing problems in *_select() on W32LRN2013-09-08
| | | | | | | | | | 1) If timeout is < 1ms, round it up to 1ms, because WaitForMultipleObjects() can't wait for time shorter than 1ms (0ms means "don't wait at all"). 2) Read data from the wakeup socket before commanding the select thread to start selecting. For some reason the socket would be in signalled state by the time winsock's select() runs, even though we emptied it the last time. So now we empty it beforehand. This should prevent us from returning early in some cases.
* -fix u/clarify/improve util/network code doxygen documentationChristian Grothoff2013-08-29
|
* -eliminate more uses of ShortHashCode, breaks signaturesChristian Grothoff2013-08-19
|
* - use SO_REUSEADDR for SOCK_STREAM type sockets onlySree Harsha Totakura2013-08-15
|
* changing time measurement from milliseconds to microsecondsChristian Grothoff2013-08-11
|
* -fix warnings on W32Christian Grothoff2013-07-24
|
* doxygenFlorian Dold2013-06-26
|
* Fix non-windows non-IPv6 build.David Barksdale2013-06-21
|
* Fixed GNUNET_CLIENT_service_test on FreeBSD.David Barksdale2013-06-20
| | | | | | | This should also fix any other platform with AF_UNIX that is not Linux or Mingw.
* Fix warnings and fix/improve logging in network.cLRN2013-03-20
|
* -check for EINPROGRESS for UNIX, hopefully fixing #2741Christian Grothoff2013-01-30
|
* W32: Remove unnecessary 0 and -1 checksLRN2012-11-01
|
* W32: optimize GNUNET_NETWORK_fdset_addLRN2012-11-01
|
* -bugfixes, code cleanupChristian Grothoff2012-10-26
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-20 03:45:19 +0000