| Age | Commit message (Expand) | Author |
|---|
| 2004-05-03 | POSIX says the length parameter for semop(2) and shmget(2) should be size_t. | Todd C. Miller |
| 2004-04-25 | change sb_mbmax to: (sb_max/MCLBYTES) * (MSIZE+MCLBYTES); ok deraadt | Markus Friedl |
| 2004-04-25 | fix typos/spelling in comments, from pedro martelletto | Jolan Luff |
| 2004-04-25 | radix tree with multipath support. from kame. deraadt ok | Jun-ichiro itojun Hagino |
| 2004-04-19 | introduce a new km_page allocator that gets pages from kernel_map using | Ted Unangst |
| 2004-04-19 | also use sbcheckreserve() for setsockopt of SO_SNDBUF and SO_RCVBUF | Theo de Raadt |
| 2004-04-19 | this is only a work in progress, we can perfect afterwards, but it is time | Theo de Raadt |
| 2004-04-17 | Don't forget to apply the M_CLUSTER flag when copying flags in m_pullup2(). | Ryan Thomas McBride |
| 2004-04-16 | not a pasto. ok millert@ | Ted Unangst |
| 2004-04-15 | rework access to emulations slightly in an effort to both merge and | Ted Unangst |
| 2004-04-13 | useless caddr_t cast removal. same md5s. | Ted Unangst |
| 2004-04-12 | make sockargs take void *, combine a len check. | Ted Unangst |
| 2004-04-11 | Crank max ptys to 992. We now have 62 pty/tty pairs for each letter | Todd C. Miller |
| 2004-04-11 | Fix calculation of the index into TTY_LETTERS. This only worked | Todd C. Miller |
| 2004-04-06 | fix siginfo generation for posted signals and avoid double ktrpsig() call for... | Michael Shalayeff |
| 2004-04-02 | rfork(RFMEM) shares complete vmspace. much more useful, and in line with | Ted Unangst |
| 2004-04-01 | use NULL for ptrs. parts from Joris Vink | Ted Unangst |
| 2004-04-01 | init kqueues normally, from pedro martelletto | Ted Unangst |
| 2004-03-20 | one proc.h is sufficient | Ted Unangst |
| 2004-03-19 | check for CLOCAL flag in right variable; from sq@oganer.net | Theo de Raadt |
| 2004-03-17 | Add missing open paren in comment noticed by jmc@ | Todd C. Miller |
| 2004-03-17 | Adjust the number of processes waiting on a semaphore properly if we're | Todd C. Miller |
| 2004-03-14 | nicer printf for copyright. from Pedro Martelletto | Ted Unangst |
| 2004-03-13 | sync comments with reality, semconfig is long gone. | Ted Unangst |
| 2004-03-12 | don't leak memory on error, from Patrick Latifi | Ted Unangst |
| 2004-03-03 | better way of finding and identifying lkms. | Ted Unangst |
| 2004-02-29 | Check for negative numbers. | Tobias Weingartner |
| 2004-02-28 | Regen | Miod Vallat |
| 2004-02-28 | Rather than adding amd64 to the 64-bit platform which do not require the | Miod Vallat |
| 2004-02-27 | Move setperf_prio to the machdep code, requested by deraadt@ | Alexander Yurchenko |
| 2004-02-27 | Introduce simple priorities mechanism so that concurrent | Alexander Yurchenko |
| 2004-02-24 | sysctl knob for bpf tunables. some tips from canacar@ | Ted Unangst |
| 2004-02-23 | Make sure we allow for the time when check_pty has not yet been called | Bob Beck |
| 2004-02-15 | add sysctl_int_arr() API similar to BSD/OS; ok itojun, deraadt, miod, henning | Markus Friedl |
| 2004-02-15 | new arg to disk_unbusy, to record separate read/write statistics. | Ted Unangst |
| 2004-02-15 | pretty code that's easier to read, same effect. | Ted Unangst |
| 2004-02-14 | Simplify hw.{cpuspeed,setperf} api moving all the sysctl stuff | Alexander Yurchenko |
| 2004-02-10 | Fix off-by-one wrt TTY_LETTERS and minor being out of bounds. Missed | Todd C. Miller |
| 2004-02-10 | Add the ptm device to pty(4). By opening /dev/ptm and using the PTMGET | Todd C. Miller |
| 2004-02-10 | Dynamically allocate space for ptys; adapted from NetBSD by beck@ | Todd C. Miller |
| 2004-02-08 | restored & repaired wcookie support; kettenis@chello.nl | Theo de Raadt |
| 2004-02-05 | Decrement reference count in error case if uvm_map(9) fails and we | Todd C. Miller |
| 2004-02-05 | Correct a reference counting bug in shmat(2); adapted from FreeBSD. | Todd C. Miller |
| 2004-02-01 | Grok union like struct in genassym.cf. Will be needed shortly. | Miod Vallat |
| 2004-01-29 | The Sxxx proc state defines start at 1, not 0. Also make the const | Todd C. Miller |
| 2004-01-28 | oh, i'll happily steal the commit. ok henning@, markus@, otto@ (iirc) | Daniel Hartmeier |
| 2004-01-26 | having the monotonic thing as DEBUG is not going to get it fixed faster, it i... | Theo de Raadt |
| 2004-01-21 | dynamic bufq support. basically a nullop. ok grange@ | Ted Unangst |
| 2004-01-21 | call timeout_startup early, so timeouts are available. report from sam smith | Ted Unangst |
| 2004-01-14 | Get rid of M_LOCKF and use pool for allocating lockf structures. | Alexander Yurchenko |
