| Age | Commit message (Expand) | Author |
|---|
| 2017-07-27 | Assert that the KERNEL_LOCK() is held prior to call csignal() and | Martin Pieuchot |
| 2017-07-24 | Extend the scope of the socket lock to protect `so_state' in connect(2). | Martin Pieuchot |
| 2017-07-20 | If pool_get() sleeps while allocating additional memory for socket | Alexander Bluhm |
| 2017-07-20 | Prepare filt_soread() to be locked. No functionnal change. | Martin Pieuchot |
| 2017-07-13 | Do not unlock the netlock in the goto out error path before it has | Alexander Bluhm |
| 2017-07-08 | Revert grabbing the socket lock in kqueue filters. | Martin Pieuchot |
| 2017-07-04 | Always hold the socket lock when calling sblock(). | Martin Pieuchot |
| 2017-07-03 | Protect `so_state', `so_error' and `so_qlen' with the socket lock in | Martin Pieuchot |
| 2017-06-27 | Add missing solock()/sounlock() dances around sbreserve(). | Martin Pieuchot |
| 2017-06-26 | Assert that the corresponding socket is locked when manipulating socket | Martin Pieuchot |
| 2017-06-20 | In ddb print socket bit field so_state in hex to match SS_ defines. | Alexander Bluhm |
| 2017-06-20 | Convert sodidle() to timeout_set_proc(9), it needs a process context | Martin Pieuchot |
| 2017-05-31 | new socketoption SO_ZEROIZE: zero out all mbufs sent over socket | Markus Friedl |
| 2017-05-27 | Push the NET_LOCK down into PF_KEY so that it can be treated like PF_ROUTE. | Claudio Jeker |
| 2017-05-15 | so_splicelen needs to be protected by the socket lock. We are now | Martin Pieuchot |
| 2017-05-15 | Enable the NET_LOCK(), take 3. | Martin Pieuchot |
| 2017-04-02 | Less convoluted code in soshutdown() | Theo de Raadt |
| 2017-03-17 | Revert the NET_LOCK() and bring back pf's contention lock for release. | Martin Pieuchot |
| 2017-03-13 | Move PRU_ATTACH out of the pr_usrreq functions into pr_attach. | Claudio Jeker |
| 2017-03-07 | Do not grab the NET_LOCK() for routing sockets operations. | Martin Pieuchot |
| 2017-03-03 | Prevent a recursion in the socket layer. | Martin Pieuchot |
| 2017-02-14 | Wrap the NET_LOCK() into a per-socket solock() that does nothing for | Martin Pieuchot |
| 2017-02-01 | In sogetopt, preallocate an mbuf to avoid using sleeping mallocs with | David Hill |
| 2017-01-27 | In sosend() the size of the control message for file descriptor | Alexander Bluhm |
| 2017-01-26 | Do not hold the netlock while pool_get() may sleep. It is not | Alexander Bluhm |
| 2017-01-25 | As NET_LOCK() is a read/write lock, it can sleep in sotask(). So | Alexander Bluhm |
| 2017-01-25 | Enable the NET_LOCK(), take 2. | Martin Pieuchot |
| 2016-12-29 | Change NET_LOCK()/NET_UNLOCK() to be simple wrappers around | Martin Pieuchot |
| 2016-12-20 | Grab the NET_LOCK() in so{s,g}etopt(), pffasttimo() and pfslowtimo(). | Martin Pieuchot |
| 2016-12-19 | Introduce the NET_LOCK() a rwlock used to serialize accesses to the parts | Martin Pieuchot |
| 2016-11-29 | m_free() and m_freem() test for NULL. Simplify callers which had their own | Jonathan Gray |
| 2016-11-23 | Some socket splicing tests on loopback hang with large mbufs and | Alexander Bluhm |
| 2016-11-22 | Enforce that pr_ctloutput is called at IPL_SOFTNET. | Martin Pieuchot |
| 2016-11-21 | Enforce that pr_usrreq functions are called at IPL_SOFTNET. | Martin Pieuchot |
| 2016-11-14 | Remove splnet() from socket kqueue code. | Martin Pieuchot |
| 2016-10-06 | Remove redundant comments that say a function must be called at | Alexander Bluhm |
| 2016-10-06 | Separate splsoftnet() from variable initialization. | Alexander Bluhm |
| 2016-09-20 | Protect soshutdown() with splsoftnet() to define one layer where | Alexander Bluhm |
| 2016-09-20 | Add some spl softnet assertions that will help us to find the right | Alexander Bluhm |
| 2016-09-15 | all pools have their ipl set via pool_setipl, so fold it into pool_init. | David Gwynne |
| 2016-09-13 | Do not raise splsoftnet() recursively in soaccept(). | Martin Pieuchot |
| 2016-09-03 | If sosend() cannot allocate a large cluster, try a small one as | Alexander Bluhm |
| 2016-09-03 | Return immediately when m_getuio() fails by invalid uio parameter. | YASUOKA Masahiko |
| 2016-08-25 | Spliced TCP sockets become faster when the output part is running | Alexander Bluhm |
| 2016-08-25 | Completely revert the M_WAIT change on the cluster allocation and | Alexander Bluhm |
| 2016-08-22 | Refactor the uio to mbuf code out of sosend and start to make use of | Claudio Jeker |
| 2016-06-13 | On localhost a user program may create a socket splicing loop. | Alexander Bluhm |
| 2016-06-12 | Fix format string in ddb show socket. | Alexander Bluhm |
| 2016-03-14 | Change a bunch of (<blah> *)0 to NULL. | Kenneth R Westerback |
| 2016-01-15 | Improve the socket panic messages further. claudio@ wants to see | Alexander Bluhm |
