summaryrefslogtreecommitdiff
path: root/sys/rump/net
AgeCommit message (Expand)Author
2014-03-03also include bsd.own.mk, so value for RUMP_NPF_TESTING can come from mk.confpooka
2014-03-03For now, make it possible to not specify -D_NPF_TESTING by compilingpooka
2014-03-03Pull in the changes from the dpdk, netmap and snabb switch repos.pooka
2014-02-26Run constructor as part of network faction bootstrap, not devicepooka
2014-02-21Fix typo in fix for PR/48606skrll
2014-02-21PR/48606: Lloyd Parkes: Drivers not using ifp->if_input but using ether_inputchristos
2014-02-14Register netisr's from component constructors instead of via a hardcodedpooka
2014-01-08OpenBSD has kqueuepooka
2014-01-02Allow kernels compiled with INET+INET6 to be booted as IPv4-only or IPv6-only.pooka
2013-12-20use _KERNEL_RCSID only in _KERNEL, _RCSID elsewherepooka
2013-12-20header massage, mostly improves portabilitypooka
2013-11-16Move rumpnet_bpfjit under MKSLJIT as well.rmind
2013-11-16Build rumpkern_sljit and rumpnet_bpfjit; include them into the sets.rmind
2013-11-16Add rumpkern_sljit and rumpnet_bpfjit modules.rmind
2013-11-11Fix syntax for undefined HAVE_GCC.joerg
2013-11-08NPF: add support for specifying the interfaces before they are attached.rmind
2013-11-07avoid gcc bug on sh3christos
2013-10-27Give the hypercall interface a version number. This is forpooka
2013-09-19NPF: G/C n-code in favour of BPF byte-code. Delete lots of code, mmm!rmind
2013-09-13Remove busversion.joerg
2013-09-13GC unused functions. Don't bother building ieee80211_acl.c, nothing injoerg
2013-08-29Remove SS_ISCONFIRMING, it is unused and TP4 will not come back.rmind
2013-08-14Partially revert previous, there are still a few gotchas inpooka
2013-08-14Separate inet and inet6, allows inet6-only rump kernels.pooka
2013-08-01Make it possible to not build virtif since it's not supportedpooka
2013-07-22In recv, align data after ether_header at _ALIGN_BYTES.pooka
2013-07-18Add librumpnet_netmpls that provides MPLS features into rump kernelskefren
2013-07-16A simple (void) is apparently too easy and traditional to makepooka
2013-07-16Explicitly ignore return value of writev() for sending a packet.pooka
2013-07-04Use kqueue on *BSD, not just NetBSD.pooka
2013-07-04Add a note explaining why there's a homegrown __STRING, __CONCAT, etc.pooka
2013-07-04Determine the interface name and the names of exported symbols basedpooka
2013-07-04Avoid not-used warning on NetBSDpooka
2013-07-04fix warningspooka
2013-07-03fix parameter namepooka
2013-07-03Adjust for rump_virtif_create() removal. Incidentally, this reducespooka
2013-07-03g/c dummyif, not useful these dayspooka
2013-07-03Allow alternate interface base names to be specified at compile time.pooka
2013-06-30Raise lo0 so that ipv6 config gets done for it.pooka
2013-06-23Make sure sockin module is initialized only oncestacktic
2013-06-14sauce previous with a commentpooka
2013-06-14Ignore too short packets. This condition is true especially for thepooka
2013-06-04On Linux, pass the inotify descriptor up so that we can correctly readpooka
2013-06-02- NPF connection tracking: rework synchronisation on tracking disable/enablermind
2013-06-01check for EOF from backing sucketpooka
2013-06-01Some minor tweaks to make this compile on Solaris.pooka
2013-06-01Assert that we get PRU_ATTACH requests only for the families wepooka
2013-06-01default portalgo to random_start, for less connections from port 65535pooka
2013-06-01cosmetic: put portalgo.c in the conceptually right placepooka
2013-06-01Add IPv6 supportstacktic