summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Expand)Author
2016-09-04Only allow bfd on host routes and non-gateway routes for now.Claudio Jeker
2016-09-04Add comments to the character-to-block device table and remove anChristian Weisgerber
2016-09-04For the allwinner,sun4i-a10-wdt disable/enable the reset-bit instead ofMarcus Glocker
2016-09-04Add PT_GNU_EH_FRAME.Mark Kettenis
2016-09-04Grab a reference to the route and free it once no longer needed.Claudio Jeker
2016-09-04Remove the IFF_LINK0 option to send the response back to the peer'sReyk Floeter
2016-09-04Bring back 2 MB socket buffers to speed up TCP. This increasesAlexander Bluhm
2016-09-04Rename local variables called 'ret' to 'err' everywhere in iwm.Stefan Sperling
2016-09-04Clean up the pile of iwm MAC context command add/send/changed functions.Stefan Sperling
2016-09-04Remove support for tape block devices. Nobody mount(8)s tapes any longer.Christian Weisgerber
2016-09-04correct the size passed to free in iatp_write_regJonathan Gray
2016-09-04Rename iwm_mvm_phy_ctxt_apply() to iwm_mvm_phy_ctxt_cmd() and get rid ofStefan Sperling
2016-09-04naddy discovered that bha no longer builds because it includes the nowTed Unangst
2016-09-04Purge routes attached to an address when this address is removed.Martin Pieuchot
2016-09-04bha depends on the deleted aha.h, so it will follow it into the atticTed Unangst
2016-09-04Remove unused iwm_mvm_binding_update() wrapper function and replace theStefan Sperling
2016-09-04Get rid of pointless iwm_mvm_time_event_send_add() wrapper.Stefan Sperling
2016-09-04Make it possible to toggle RTF_BFD via RTM_CHANGE and fix some minor thingsClaudio Jeker
2016-09-04Introduce Dynamic Profiling, a ddb(4) based & gprof compatible kernelMartin Pieuchot
2016-09-04TCB_GET_MEMBER() is no longer used after the TIB changesPhilip Guenther
2016-09-04Restrict MSR access to supported ones, log invalid accesses.Mike Larkin
2016-09-04Telling gcc the TCB pointer is in %r2 via the 'register asm' extensionPhilip Guenther
2016-09-04ld also knows about PT_OPENBSD_BOOTDATAPhilip Guenther
2016-09-04Do "goto failed" in case returning EAGAIN as well.YASUOKA Masahiko
2016-09-04Modify to ANSI-style function declarations. No binary change.Kenji Aoyama
2016-09-03ld knows PT_OPENBSD_RANDOMIZE nowTheo de Raadt
2016-09-03handle the remaining bits of the packet we are sentPeter Hessler
2016-09-03only clear the error counter if we are successful, not if we getPeter Hessler
2016-09-03aha remnants lurking in hereTed Unangst
2016-09-03make intrace a volatile, while the compiler does the right thing stillJasper Lievisse Adriaanse
2016-09-03the src tree is growing too large. delete the aha and ahb drivers inTed Unangst
2016-09-03zap eyesore whitespace at EOLJasper Lievisse Adriaanse
2016-09-03handle timer limitationsPeter Hessler
2016-09-03we really should ntohl fields coming in from the networkPeter Hessler
2016-09-03remove strongarm files that were abused by zaurusJonathan Gray
2016-09-03jsg@ also found some uninitialized variables, thank you!Peter Hessler
2016-09-03in order to avoid problems with precise timing, obey rfc 5880 section 6.8.7,Peter Hessler
2016-09-03if_get requires the corresponding if_put call.Peter Hessler
2016-09-03Only use uaddr_exe for address selection when PROT_EXEC is requestedStefan Kempf
2016-09-03Fixed missing null check in switchctl.cKazuya Goda
2016-09-03Remove iwm_release() by inlining its implementation into its only caller.Stefan Sperling
2016-09-03Switch back to 256 KB socket buffer size for TCP. With 2 MB theAlexander Bluhm
2016-09-03iwm was undecided between 'if (err != 0)' and 'if (err)'; switch to 'if (err)'Stefan Sperling
2016-09-03Let purge thread to remove once rules, not packets.Alexandr Nedvedicky
2016-09-03In iwm, move assignments to 'err' outside of if-statements.Stefan Sperling
2016-09-03remove testing printf that snuck in accidentally.Peter Hessler
2016-09-03instead of doing everything manually, use goto cleanup idiomPeter Hessler
2016-09-03check the TTL field for incoming packets, we only support directlyPeter Hessler
2016-09-03The iwm code was torn between 'error' and 'err'; error -> err everywhereStefan Sperling
2016-09-03remove cnw driver. it was never enabled anywhere except on amd64 (and onlyTed Unangst