| Age | Commit message (Expand) | Author |
|---|
| 2015-11-04 | Basic framework to test the rtable_* API directly from userland. | Martin Pieuchot |
| 2015-11-04 | The "Lenovo Thinkpad USB 3.0 Ethernet Adapter" works with cdce(4). | Stefan Sperling |
| 2015-11-04 | clean up HISTORY; | Jason McIntyre |
| 2015-11-04 | fix a memory leak in multiple error paths | Jonathan Gray |
| 2015-11-04 | use mq_purge to free a queue of mbufs. | David Gwynne |
| 2015-11-04 | replace while (ml_dequeue()) m_freem(); with ml_purge(); | David Gwynne |
| 2015-11-04 | use ml_purge to flush the arp hold lists instead of dequeue/free loops. | David Gwynne |
| 2015-11-04 | use IFQ_IS_EMPTY to see if if_snd is empty, not IF_IS_EMPTY | David Gwynne |
| 2015-11-04 | replace the ifqueues used for tulip_txq and _rxq with mbuf_lists. | David Gwynne |
| 2015-11-03 | - fixes potential use-after-free in pfr_set_addrs() | Alexandr Nedvedicky |
| 2015-11-03 | Remove dead assignments. | Charles Longeau |
| 2015-11-03 | Remove dead assignment and newly created unused variable. | Charles Longeau |
| 2015-11-03 | Disable TCP/UDP TX hardware checksumming if an IPv4 packet contains | Christian Weisgerber |
| 2015-11-03 | If pledge(2) fails use perror instead of log_it(RealUser, ...) since | Todd C. Miller |
| 2015-11-03 | AF_UNIX connect is a "unix" operation, not "rpath wpath" | Theo de Raadt |
| 2015-11-03 | improve the load balancing text; ok mpi | Jason McIntyre |
| 2015-11-03 | Improve explanations throughout | Theo de Raadt |
| 2015-11-03 | It is only necessary to swap the effective gid when reading a file. | Todd C. Miller |
| 2015-11-03 | It is only necessary to swap the effective gid when reading a file. | Todd C. Miller |
| 2015-11-03 | There really is no double standard. Treat m4 just like openssl, by | Theo de Raadt |
| 2015-11-03 | pledge_aftersyscall has been reduced to one case, "getpw", to open a | Theo de Raadt |
| 2015-11-03 | Ignore disks whose media is not present. | YASUOKA Masahiko |
| 2015-11-03 | Recognize carp(4) interfaces (yes, they are special...) in order to send | Martin Pieuchot |
| 2015-11-03 | Detach the client we are looping over, from Thomas Adam. | Nicholas Marriott |
| 2015-11-03 | Don't allow the user to enter GPT partition names too large to fit | Kenneth R Westerback |
| 2015-11-03 | Add & use a #define GPTPARTNAMESIZE rather than the magic number | Kenneth R Westerback |
| 2015-11-03 | Do not clear M_PROTO1 flag before calling if_start() because pseudo- | Martin Pieuchot |
| 2015-11-03 | Missing comma. | Martin Pieuchot |
| 2015-11-03 | use an mbuf_queue for the svcq instead of an ifqueue. | David Gwynne |
| 2015-11-03 | pull packets off the send queue with IFQ_DEQUEUE instead of IF_DEQUEUE. | David Gwynne |
| 2015-11-03 | fix a memory leak in an error path | Jonathan Gray |
| 2015-11-03 | we need an unbreakable space in O(log2 n). | Igor Sobrado |
| 2015-11-03 | regress: pledge | Sebastien Marie |
| 2015-11-03 | Use "Loaded Image Protocol" instead of "Loaded Device Path Protocol" | YASUOKA Masahiko |
| 2015-11-03 | regress: pledge | Sebastien Marie |
| 2015-11-03 | regress pledge: check the regress with systrace enabled too | Sebastien Marie |
| 2015-11-03 | When updating an interface, do not delete the kif_node but let | Gerhard Roth |
| 2015-11-03 | zap the correspoonding note in systrace(4) - tedu says | Jason McIntyre |
| 2015-11-03 | Delete VBITS: it's unused and the definition is actually wrong | Philip Guenther |
| 2015-11-03 | Blank line after pledge call. | mmcc |
| 2015-11-03 | Blank line after pledge call. | mmcc |
| 2015-11-03 | Blank lines around pledge call. | mmcc |
| 2015-11-03 | Blank line after pledge call. | mmcc |
| 2015-11-03 | Surround pledge call with blank lines. | mmcc |
| 2015-11-03 | Surround pledge calls with blank lines. | mmcc |
| 2015-11-03 | Blank line after pledge(). | mmcc |
| 2015-11-03 | Blank line before pledge(). | mmcc |
| 2015-11-03 | Add blank line after pledge call. | mmcc |
| 2015-11-03 | Add blank lines surrounding pledge call, as per decree of deraadt@ | mmcc |
| 2015-11-03 | Use send(MSG_NOSIGNAL) instead of mucking with the disposition of SIGPIPE | Philip Guenther |
