summaryrefslogtreecommitdiff
path: root/network_io/os2
Commit message (Expand)AuthorAgeFilesLines
* OS/2: Add unix inet_ntop() to OS/2 build.Brian Havard2000-11-252-0/+2
* Rename sa_len to salen in apr_sockaddr_t. Some platforms have a macroRyan Bloom2000-11-211-7/+7
* apr_create_tcp_socket() has been removed. Use apr_create_socket()Jeff Trawick2000-11-211-5/+0
* Change apr_bind() to take apr_sockaddr_t.Jeff Trawick2000-11-191-4/+6
* APR: Change apr_connect() to take apr_sockaddr_t instead of hostname.Jeff Trawick2000-11-171-22/+4
* OS/2: fix some minor breakage after latest changes to addresses.Brian Havard2000-11-171-1/+2
* Get OS/2 closer to building with updated network_io API.Jeff Trawick2000-11-161-2/+24
* port initial set of David Reid's IPv6 support to OS/2Jeff Trawick2000-11-162-33/+63
* OS/2 changes to the way dependencies are built in APR makefilesJeff Trawick2000-11-151-75/+2
* OS/2: Convert apr_get_remote_hostname() to apr_get_hostname(), supporting bothBrian Havard2000-11-141-5/+11
* OS/2: soclose() segfaults if passed -1 so prevent this from happening ifBrian Havard2000-11-141-0/+5
* OS/2: return APR_EOF from apr_read() when there's no more data. This matchesBrian Havard2000-11-141-1/+1
* Fix OS/2 build after the move of private APR headers to the include/arch/Brian Havard2000-11-141-40/+64
* Now lives in arch/os2/Brian Havard2000-11-141-97/+0
* Fix arg type to match prototype.Brian Havard2000-11-131-1/+1
* Rework apr_poll to use APR time types, fixing bug where an indefiniteBrian Havard2000-11-131-10/+3
* Convert a lot of apr_ssize_t to apr_size_t. We don't ever accept or returnRyan Bloom2000-11-071-2/+2
* Make 2nd parm of apr_connect() const char * instead of char *.Jeff Trawick2000-11-021-1/+1
* Sort out OS/2 APR error handling with new canonical test macro.Brian Havard2000-10-092-48/+52
* APRize disabling nagle (setting TCP_NODELAY).Brian Havard2000-08-221-0/+5
* Include apr_strings.h in a bunch more places that need it.Brian Havard2000-08-061-0/+1
* Remaining cleanup of ap_ -> apr_ and AP_ -> APR_ transformation...William A. Rowe Jr2000-08-062-2/+2
* prefix libapr functions and types with apr_Doug MacEachern2000-08-025-56/+56
* a while back, we said "segfault on invalid params rather than returning anGreg Stein2000-07-081-17/+2
* const-ify the userdata interfacesGreg Stein2000-07-081-2/+3
* OS/2: Add ap_getsocketopt() implementation.Brian Havard2000-07-011-0/+14
* Remove const from ap_socket_t parameters on APR functions so thatJeff Trawick2000-06-261-1/+1
* Make `mm' an INCDIR in shmem/unix, so that the INCDIR rewriter canSascha Schumann2000-04-301-1/+0
* Update comment for depend target. The template makefiles are now calledSascha Schumann2000-04-301-1/+1
* back out APR_ prefix for TRUE,FALSEDoug MacEachern2000-04-281-2/+2
* prefix TRUE,FALSE with APR_Doug MacEachern2000-04-281-2/+2
* OS/2: Fix up ap_poll stuff after recent changes to ap_remove_poll_socket()Brian Havard2000-04-281-4/+9
* OS/2: Fix some uninitialized fields in accepted socketsBrian Havard2000-04-241-0/+2
* Move apr_config.h and apr_config.hw to apr_private.h and apr_private.hw.Ryan Bloom2000-04-221-1/+1
* APR_ENOCONT doesn't make any sense without contexts. Replaced them all withRyan Bloom2000-04-191-1/+1
* APR_SO_TIMEOUT now takes microseconds instead of seconds. (The newJeff Trawick2000-04-162-4/+4
* Convert ap_poll() to use microseconds instead of seconds forJeff Trawick2000-04-161-2/+5
* Change ap_context_t to ap_pool_t. This compiles, runs, and serves pagesRyan Bloom2000-04-144-7/+7
* First stage of migrating OS/2 code to new error code mappings, define & useBrian Havard2000-04-086-25/+23
* socket_cleanup() is local & not namespace protected, should be static.Brian Havard2000-04-041-1/+1
* Finish the APR naming cleanup. This removes the struct's fromRyan Bloom2000-04-034-28/+28
* fix the rest of the struct foo_t to struct ap_foo_t. Next is removing theRyan Bloom2000-04-035-30/+30
* Update to Apache Software License version 1.1Roy T. Fielding2000-03-317-225/+214
* Update some dependencies.Brian Havard2000-03-161-22/+30
* Fix all the License issues. Including:Ryan Bloom2000-03-107-63/+63
* OS/2: ap_send/ap_recv/ap_sendv should set len to 0, not -1, when an errorBrian Havard2000-02-141-6/+6
* OS/2: Rework ap_send/ap_recv to time out properly on blocking sockets.Brian Havard1999-11-034-80/+74
* Fix an APR socket bug. ap_bind would always bind to INADDR_ANY, noManoj Kasichainula1999-11-021-1/+0
* We were passing NULL to ap_register_cleanup, breaking mod_cgi. ReplaceManoj Kasichainula1999-11-011-2/+2
* OS/2: Bypass EMX library's socket code and go straight to the OS/2 TCP/IPBrian Havard1999-10-298-102/+444