summaryrefslogtreecommitdiffstats
path: root/usr.sbin/bind/lib/isc/unix/socket.c (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Finish moving of dig(1) to /usr/bin/dig by removing the sources in theflorian2020-02-111-2753/+0
| | | | | old location. dig(1) sources can be found in src/usr.bin/dig. ok deraadt@
* Support for queries on ports != 53jca2020-01-261-2/+2
| | | | | | | | | | | | | | - dig(1) -p now works, as already supported by dig(1) from ports/net/isc-bind thanks to sthen@ - "set port=..." also works in nslookup(1) - host(1) has no support for alternate destination ports This adds "inet" to the pledge promises of all three programs, a compromise deemed reasonable and suggested by deraadt@; instead of slightly more complicated patches by myself and florian@ to avoid "inet" in the general case. Suggested by deraadt@, ok florian@ sthen@
* Remove unused code from socket.cflorian2020-01-261-590/+18
| | | | OK tedu
* get rid of lib/isc/nothreadsflorian2020-01-261-3/+0
| | | | OK tedu
* Oops, one break too many removed; unbreak +tcp.florian2020-01-241-0/+1
|
* Remove overly complicated set_rcvbuf. Very likely not needed in dig et al.florian2020-01-231-68/+0
| | | | OK deraadt
* Remove socket types that we are not going to use, like AF_UNIX sockets.florian2020-01-231-1183/+30
| | | | OK deraadt
* config.h and platform.h are unused nowflorian2020-01-221-2/+2
|
* unifdef the last config.h bitsflorian2020-01-221-54/+0
| | | | OK millert
* Spell out PORT_NONBLOCK as O_NONBLOCK and ISC_SOCKADDR_LEN_T asflorian2020-01-221-15/+12
| | | | | socklen_t. OK millert
* eliminate many other possibilities in favor of just using select() here.tedu2020-01-221-424/+0
| | | | ok florian
* remove mutex.h etc. since this code is now unthreaded.tedu2020-01-211-238/+0
| | | | ok florian
* Rip out custom memory allocator.florian2020-01-201-85/+59
| | | | OK otto
* As millert@ points out we can replace isc_string_touint64 withflorian2020-01-201-1/+1
| | | | | strtoull(3) and then get rid of lib/isc/string.c OK millert
* Do not use a message catalog for error messages.florian2020-01-201-129/+32
| | | | | It's not installed anyway. OK millert
* unifdef threading related definesflorian2020-01-201-19/+0
| | | | input & OK millert
* We are not using json, xml nor httpd.florian2020-01-201-279/+2
| | | | OK millert
* Remove various unused bits and pieces from lib/isc.florian2020-01-091-654/+0
| | | | | | | Minus 3k lines. Input deraadt OK millert
* Use normal int types like int32_t instead of isc_int32_t.florian2020-01-091-5/+5
| | | | OK millert
* Nuke atomics support. This gets rid of 4k lines of code including a bunchflorian2020-01-091-186/+1
| | | | | | | of assembly code. It's used for various statistics gathering in the bind daemon which we won't need in dig etc. OK millert
* Assorted unifdefs:florian2020-01-091-17/+0
| | | | | | | | | | | | | | | | | | | | #define ISC_PLATFORM_NEEDSYSSELECTH 1 #define ISC_PLATFORM_HAVESTATNSEC 1 #define ISC_PLATFORM_RLIMITTYPE rlim_t #define ISC_PLATFORM_HAVELONGLONG 1 #undef ISC_PLATFORM_BRACEPTHREADONCEINIT #undef ISC_PLATFORM_USEDECLSPEC #define ISC_PLATFORM_HAVESYSUNH 1 #define ISC_PLATFORM_HAVESTRINGSH 1 #define ISC_PLATFORM_WANTAES 1 #define ISC_PLATFORM_USESIT 1 #define LIBISC_EXTERNAL_DATA #define LIBDNS_EXTERNAL_DATA #define LIBISCCFG_EXTERNAL_DATA #define LIBBIND9_EXTERNAL_DATA #define LIBTESTS_EXTERNAL_DATA prodding deraadt@
* unifdef printf and string functions:florian2020-01-091-1/+1
| | | | | | | | | | | | | | #undef ISC_PLATFORM_NEEDVSNPRINTF #undef ISC_PLATFORM_NEEDSPRINTF #undef ISC_PLATFORM_NEEDPRINTF #undef ISC_PLATFORM_NEEDFPRINTF #define ISC_PLATFORM_QUADFORMAT "ll" #undef ISC_PLATFORM_NEEDSTRSEP #undef ISC_PLATFORM_NEEDSTRLCPY #undef ISC_PLATFORM_NEEDSTRLCAT #undef ISC_PLATFORM_NEEDSTRTOUL #undef ISC_PLATFORM_NEEDMEMMOVE #undef ISC_PLATFORM_NEEDSTRCASESTR
* unifdef network defines:florian2020-01-091-126/+0
| | | | | | | | | | | | | | | | | | | | | | | | | #define ISC_PLATFORM_HAVESALEN 1 #define ISC_PLATFORM_HAVEIPV6 1 #undef ISC_PLATFORM_NEEDIN6ADDRANY #undef ISC_PLATFORM_NEEDIN6ADDRLOOPBACK #define ISC_PLATFORM_HAVEIN6PKTINFO 1 #undef ISC_PLATFORM_HAVEINADDR6 #define ISC_PLATFORM_HAVESCOPEID 1 #undef ISC_PLATFORM_NEEDNTOP #undef ISC_PLATFORM_NEEDPTON #undef ISC_PLATFORM_NEEDPORTT #undef ISC_PLATFORM_HAVELIFCONF #undef ISC_PLATFORM_HAVEIF_LADDRCONF #undef ISC_PLATFORM_HAVEIF_LADDRREQ #define ISC_NET_BSD44MSGHDR 1 #define ISC_PLATFORM_HAVEIFNAMETOINDEX 1 #undef ISC_PLATFORM_FIXIN6ISADDR #define ISC_PLATFORM_HAVESOCKADDRSTORAGE 1 #define ISC_PLATFORM_HAVEKQUEUE 1 #undef ISC_PLATFORM_HAVEEPOLL #undef ISC_PLATFORM_HAVEDEVPOLL prodding deraadt@
* Remove thread support. Minus 4k lines.florian2020-01-071-42/+0
| | | | OK millert
* unifdef _AIXsthen2019-12-181-34/+0
|
* Add back some SOCK_DNS to socket() calls, and add a few more whilederaadt2019-12-171-3/+3
| | | | here (since this codebase will only support dig & nslookup)
* update to 9.10.8-P1, last isc-licensed releasesthen2019-12-171-129/+104
|
* Update to bind-9.10.5-P3, which appears to have been the last ISC version.deraadt2019-12-161-761/+3427
| | | | | | | | | We only use this tree to build dig and nslookup. Our previous version predated edns0 support in those tools, and we want that. This is the worst code I've looked at in years, with layers and layers of spaghetti abstraction clearly unfit for reuse, but then reused anyways, and the old ones remain behind. So this is a 8MB diff. florian, sthen, and otto tried this merge before but failed.
* Change last non-/gnu/ fcntl(x, F_GETFL, 0) strays to fcntl(x, F_GETFL).krw2016-04-161-1/+1
| | | | | | No functional change. ok millert@
* pledge bind(1), dig(1), and nslookup(1). This modifies the underlyingderaadt2015-10-251-2/+2
| | | | | | | ISC library to use socket(2) with the SOCK_DNS flag. As a result, the port commands are disabled (such practice is rare in the wild these days, and pretty incompatible with the DNS vs regular socket concept in pledge..) ok bluhm phessler reyk, etc
* Update to BIND 9.4.2-P2 and adapt our dynamic select changes. OK deraadt@millert2008-08-041-134/+302
|
* Dynamically allocate fd_sets so we are not limited to FD_SETSIZEmillert2008-07-161-49/+137
| | | | connections. With help from djm@. OK djm@
* msg_controllen has to be CMSG_SPACE so that the kernel can account forderaadt2008-03-241-27/+15
| | | | | | | each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This works now that kernel fd passing has been fixed to accept a bit of sloppiness because of this ABI repair. lots of discussion with kettenis
* CMSG_LEN and CMSG_SPACE are not neccessarily the same size. So if youderaadt2008-03-191-15/+27
| | | | | | are going to allocate/free a block of memory for the message contents, you have to calculate and track the sizes seperately. This change is just like all the other CMSG changes...
* more strcpy fixesjakob2007-12-091-3/+3
|
* resolve conflictsjakob2007-12-091-51/+339
|
* resolve conflicts and regen configure scriptjakob2007-01-101-5/+27
|
* resolve conflictsjakob2006-04-051-41/+64
|
* resolve conflictsjakob2005-03-221-11/+9
|
* resolve conflictsjakob2004-09-281-126/+245
|
* Privilege seperation for named. Allows named to handle address/interfacecanacar2004-03-121-1/+12
| | | | | | | | | changes without restart. If you use non-standard ports in named configuration make sure they are > 1024. Also /var/named/etc/rndc.key (if any) must be readable by group named. Initial work and testing by itojun@, jakob@, hints, help from henning@, avsm@, beck@. ok henning, beck, avsm, deraadt
* update to BIND v9.2.3. ok todd@jakob2003-11-181-5/+34
|
* update to BIND 9.2.2-release. ok millert@jakob2003-03-051-7/+7
|
* ISC BIND version 9.2.2rc1jakob2003-01-201-0/+3357
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.