summaryrefslogtreecommitdiff
path: root/sys/netinet/in_pcb.c
AgeCommit message (Collapse)Author
2014-07-12Resize the pcb hashtable automatically. The table size will be doubledYASUOKA Masahiko
when the number of the hash entries reaches 75% of the table size. ok dlg henning, 'commit in' claudio
2014-06-04Only use ifa_ifwithaddr() to check if the binding address is on theMartin Pieuchot
system. Yes, this is ugly for the moment because OpenBSD prevents you from binding a tcp socket to broadcast address and checking for a broadcast address is... funny! If you've ever wondered why would lead people to write: ina.s_addr != ia->ia_addr.sin_addr.s_addr instead of: ina.s_addr == ia->ia_broadaddr.sin_addr.s_addr Well this is because all the IPv4 addresses belonging to your lo(4) interfaces match the second idiom. Hopefully we'll get rid of this hack soon. ok jca@, mikeb@
2014-05-07Remove the last hacks concerning the global list of IPv4 addresses in theMartin Pieuchot
source address selection logic. These hacks were only relevant for the NFS diskless boot code in order to pick the local broadcast address of the only configured interface. So, be explicit and set this address directly. Tested by florian@, ok henning@, beck@, chrisz@
2014-04-18Invert the signature logic of in{,6}_selectsrc, make them return theJeremie Courreges-Anglas
error code and pass the resulting source address back to the caller through a pointer, as suggested by chrisz. This gives us more readable code, and eases the deletion of useless checks in the callers' error path. Add a bunch of "0 -> NULL" conversions, while here. ok chrisz@ mpi@
2014-04-16Merge in_fixaddr() into in_selectsrc() in order to prepare forMartin Pieuchot
IP_SENDSRCADDR support. This reduces the differences with the IPv6 version and kill some comments that are no longer true. ok jca@, chrisz@, mikeb@
2014-04-07Retire kernel support for SO_DONTROUTE, this time without breakingMartin Pieuchot
localhost connections. The plan is to always use the routing table for addresses and routes resolutions, so there is no future for an option that wants to bypass it. This option has never been implemented for IPv6 anyway, so let's just remove the IPv4 bits that you weren't aware of. Tested a least by lteo@, guenther@ and chrisz@, ok mikeb@, benno@
2014-04-06factor out source and destination address mangling from in_pcbconnect()Christopher Zimmermann
for later reuse in udp_output(). "Apart from that OK" claudio@
2014-04-06Remove redundant call to in{,6}_pcbbind() from tcp PRU_CONNECT.Christopher Zimmermann
Make sure that in_pcbbind() is called from in_pcbconnect() by KASSERTing that local port == 0 implies an unspecified local address. OK claudio@
2014-03-28revert "Retire kernel support for SO_DONTROUTE" diff, which does bad thingsStuart Henderson
for localhost connections. discussed with deraadt@
2014-03-27Retire kernel support for SO_DONTROUTE, since the plan is to alwaysMartin Pieuchot
use the routing table there's no future for an option that wants to bypass it. This option has never been implemented for IPv6 anyway, so let's just remove the IPv4 bits that you weren't aware of. Tested by florian@, man pages inputs from jmc@, ok benno@
2014-01-18There was an awkward behaviour after a connection had been divertedAlexander Bluhm
to a socket. When the application removed the socket, the pf state would persist. A new connection did not hit the divert rule as the state still grabed the packet. The solution is to unlink the associated divert state when the socket gets destroyed. This is possible as both are linked together and a divert state without socket does not make sense. OK mikeb@
2013-12-20Switch inpt_queue from CIRCLEQ to TAILQ. Thus ending use of CIRCLEQKenneth R Westerback
in the base. Ports fixes to follow shortly for the two ports (gkrellm and net-snmp) affected. ok zhuk@ millert@
2013-10-23Back when some NRL code was merged into KAME to create the *BSD IPV6Theo de Raadt
stack (factoid: by a bunch of people in my living room), some compatibility #define's were created to shim incompatible inpcb access methods. There was an understanding they would eventually be removed. Since they are error prone, and 1999 is a long time ago, now they die. ok mikeb claudio mpi
2013-10-20Make the IPv4 and IPv6 code look similar to allow easy comparison.Alexander Bluhm
While there fix some white space errors. No functional change. OK claudio@ phessler@
2013-10-20Put a large chunk of the IPv6 rdomain support in-tree.Peter Hessler
Still some important missing pieces, and this is not yet enabled. OK bluhm@
2013-10-18In in_pcblookup() convert a hand crafted loop into a LIST_FOREACHAlexander Bluhm
macro. No binary change. OK mpi@
2013-10-17The header file netinet/in_var.h included netinet6/in6_var.h. ThisAlexander Bluhm
created a bunch of useless dependencies. Remove this implicit inclusion and do an explicit #include <netinet6/in6_var.h> when it is needed. OK mpi@ henning@
2013-09-06Make use of IFP_TO_IA() instead of rolling our own copy.Martin Pieuchot
ok bluhm@, henning@
2013-06-01Remove redundant call to splsoftnet(). There is aAlexander Bluhm
splsoftassert(IPL_SOFTNET) at the start of in_pcbdetach(). From David Hill
2013-05-31The function rip6_ctlinput() claims that sa6_src is constant toAlexander Bluhm
allow the assingment of &sa6_any. But rip6_ctlinput() could not guarantee that as it casted away the const attribute when it passes the pointer to in6_pcbnotify(). Replace sockaddr with const sockaddr_in6 in the in6_pcbnotify() parameters. This reduces the number of casts. Also adjust in6_pcbhashlookup() to handle the const attribute correctly. Input and OK claudio@
2013-04-09Remove read-only ipsec variables and directly use defines instead.Martin Pieuchot
ok mikeb@, markus@
2013-04-02Use macros sotoinpcb() and intotcpcb() instead of casts. Use NULLAlexander Bluhm
instead of 0 for pointers. No binary change. OK mpi@
2013-03-31The call to in_pcballoc() in user request attach was handled inAlexander Bluhm
three different ways. Use the same code in udp_usrreq() and rip_usrreq() and rip6_usrreq(). This also fixes a pcb and socket leak in udp_usrreq() in case soreserve() fails. Put an splsoftassert() into in_pcballoc() for safety. OK mpi@
2013-03-29Declare struct pf_state_key in the mbuf and in_pcb header files toAlexander Bluhm
avoid ugly casts. OK krw@ tedu@
2013-03-28Replace some casts to struct in_ifaddr pointer by ifatoia() or NULL.Martin Pieuchot
ok millert@, haesbaert@, bluhm@
2013-03-28Unfortunately the satosin, sintosa, ifatoia, satosin6, sin6tosa,Alexander Bluhm
ifatoia6 macros do not check the source type. They just cast anything. Remove needless casts and do not use those macros if the source type does not match. Remove duplicate defines. No binary change. OK kettenis@ krw@
2013-02-05Make the inet6 and inet cases in in_pcbrtentry() look similar.Alexander Bluhm
No functional change. OK millert@
2013-02-04Set sa_len correctly for PF_INET (and be consistent with the PF_INET6Todd C. Miller
case). On our platforms, sizeof(struct sockaddr) matches sizeof(struct sockaddr_in) so this hasn't caused problems but it is still incorrect. From Anthony Low. OK martynas@ bluhm@
2013-01-16Pass struct inpcb pointer to in_pcb...() functions instead of voidAlexander Bluhm
pointer. Allows stricter type checking. No functional change. OK claudio@
2012-09-20spltdb() was really just #define'd to be splsoftnet(); replace the formerBret Lambert
with the latter no change in md5 checksum of generated files ok claudio@ henning@
2012-07-12Be way more careful when accessing a possibly cached route in_selectsrc()Claudio Jeker
since it may already been gone. Fixes panic seen by stsp@ when unplugging a used USB interface. Tested and OK stsp@
2012-03-06Check if route is still valid when getting the cached rt entry of a pcb.Claudio Jeker
While there make sure we do the lookup in the correct routing table. OK mikeb, henning and phessler
2012-01-11Put an splsoftassert(IPL_SOFTNET) into in_pcbdetach().Alexander Bluhm
ok mikeb@
2011-07-06allow /31s on broadcast interfaces (eg ethernet) to work as per rfc3021.David Gwynne
the issue in our kernel was the broadcast address calculated on the /31 caused a ton of checks for use of broadcast addresses to kick in and prevent one of the two addresses on the /31 from being used. this diff basically detects if a /31 has been configured and doesnt configure a broadcast address for it, which makes the ips usable for normal traffic. i wrote this so i could interoperate with "carrier" network gear better, and sthen wants it so he can conserve address space use. the further special casing of broadcast address handling was from claudio@ ok claudio@ markus@ sthen@ henning@
2011-07-05ansifyDavid Hill
ok claudio@
2011-07-04Bye bye pf_test6(). Only one pf_test function for both IPv4 and v6.Claudio Jeker
The functions were 95% identical anyway. While there use struct pf_addr in struct pf_divert instead of some union which is the same. OK bluhm@ mcbride@ and most probably henning@ as well
2011-05-29Move the source address selection for multicast destinations thatClaudio Jeker
specify the outgoing interface with a multicast option up. Doing this before the route lookup allows multicast traffic to work even when the default 224/4 reject route is installed. Raw IP and IPv6 already had this behaviour. Based on work by Christiano F. Haesbaert. OK sthen@
2011-05-13Revert the pf->socket linking diff.Owain Ainsworth
at least krw@, pirofti@ and todd@ have been seeing panics (todd and krw with xxxterm not sure about pirofti) involving pool corruption while using this commit. krw and todd confirm that this backout fixes the problem. ok blambert@ krw@, todd@ henning@ and kettenis@ Double link between pf states and sockets. Henning has already implemented half of it. The additional part is: - The pf state lookup for outgoing packets is optimized by using mbuf->inp->state. - For incomming tcp, udp, raw, raw6 packets the socket lookup always is optimized by using mbuf->state->inp. - All protocols establish the link for incomming packets. - All protocols set the inp in the mbuf for outgoing packets. This allows the linkage beginning with the first packet for outgoing connections. - In case of divert states, delete the state when the socket closes. Otherwise new connections could match on old states instead of being diverted to the listen socket. ok henning@
2011-04-28Make in_broadcast() rdomain aware. Mostly mechanical change.Claudio Jeker
This fixes the problem of binding sockets to broadcast IPs in other rdomains. OK henning@
2011-04-28Allow SOCK_DGRAM sockets to be bound to the local network broadcast addr.Claudio Jeker
These sockets will only see broadcast sent to that particular address. Some applications like samba use this feature. Problem found and initial version by sthen@, OK sthen@, henning@
2011-04-24Double link between pf states and sockets. Henning has alreadyAlexander Bluhm
implemented half of it. The additional part is: - The pf state lookup for outgoing packets is optimized by using mbuf->inp->state. - For incomming tcp, udp, raw, raw6 packets the socket lookup always is optimized by using mbuf->state->inp. - All protocols establish the link for incomming packets. - All protocols set the inp in the mbuf for outgoing packets. This allows the linkage beginning with the first packet for outgoing connections. - In case of divert states, delete the state when the socket closes. Otherwise new connections could match on old states instead of being diverted to the listen socket. ok henning@
2011-04-19reintroduce using the RB tree for local address lookups. this isDavid Gwynne
confusing because both addresses and broadcast addresses are put into the tree. there are two types of local address lookup. the first is when the socket layer wants a local address, the second is in ip_input when the kernel is figuring out the packet is for it to process or forward. ip_input considers local addresses and broadcast addresses as local, however, the handling of broadcast addresses is different depending on whether ip_directedbcast is set. if if ip_directbcast is unset then a packet coming in on any interface to any of the systems broadcast addresses is considered local, otherwise the broadcast packet must exist on the interface it was received on. the code also needs to consider classful broadcast addresses so we can continue some legacy applications (eg, netbooting old sparcs that use rarp and bootparam requests to classful broadcast addresses as per PR6382). this diff maintains that support, but restricts it to packets that are broadcast on the link layer (eg, ethernet broadcasted packets), and it only looks up addresses on the local interface. we now only support classful broadcast addresses on local interfaces to avoid weird side effects with packets routed to us. the ip4 socket layer does lookups for local addresses with a wrapper around the global address tree that rejects matches against broadcast addresses. we now no longer support bind sockets to broadcast addresses, no matter what the value of ip_directedbcast is. ok henning@ testing (and possibly ok) claudio@
2011-04-14Backout the in_iawithaddr() -> ifa_ifwithaddr() change.Claudio Jeker
There is a massive issue with broadcast addrs because ifa_ifwithaddr() handles them differently then in_iawithaddr().
2011-04-04in_iawithaddr -> ifawithaddrHenning Brauer
the latter is now much faster and the former becomes a wrapper + compat hacks around the latter in a bit. ok dlg sthen and "ah hai" from various in various bars here
2010-07-03Fix the naming of interfaces and variables for rdomains and rtablesPhilip Guenthe
and make it possible to bind sockets (including listening sockets!) to rtables and not just rdomains. This changes the name of the system calls, socket option, and ioctl. After building with this you should remove the files /usr/share/man/cat2/[gs]etrdomain.0. Since this removes the existing [gs]etrdomain() system calls, the libc major is bumped. Written by claudio@, criticized^Wcritiqued by me
2010-05-07Start cleaning up the mess called rtalloc*. Kill rtalloc2, make rtalloc1Claudio Jeker
accept flags for report and nocloning. Move the rtableid into struct route (with a minor twist for now) and make a few more codepathes rdomain aware. Appart from the pf.c and route.c bits the diff is mostly mechanical. More to come... OK michele, henning
2010-01-15Replace pool_get() + bzero() with pool_get(..., PR_ZERO).Charles Longeau
With input from oga@ and krw@ ok oga@ krw@ thib@ markus@ mk@
2009-12-23The process's rdomain should be, well, per-process and not per-rthread,Philip Guenthe
so put it in struct process instead of struct proc. While at it, move the p_emul member inside struct proc so that it gets copied automatically instead of requiring manual assignment. ok deraadt@
2009-11-27Add setrdomain() and getrdomain() system calls. Committing now toPhilip Guenthe
catch the libc major bump per request from deraadt@ Diff by reyk. ok guenther@
2009-11-13Extend the protosw pr_ctlinput function to include the rdomain. This isClaudio Jeker
needed so that the route and inp lookups done in TCP and UDP know where to look. Additionally in_pcbnotifyall() and tcp_respond() got a rdomain argument as well for similar reasons. With this tcp seems to be now fully rdomain save and no longer leaks single packets into the main domain. Looks good markus@, henning@
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-19 19:09:56 +0000