| Age | Commit message (Expand) | Author | Files | Lines |
|---|
| 2020-01-21 | Import dt(4) a driver and framework for Dynamic Profiling. |   mpi | 1 | -1/+12 |
| 2019-11-02 | softclock: move softintr registration/scheduling into timeout module |   cheloha | 1 | -12/+3 |
| 2019-08-02 | per-process itimers: itimerval -> itimerspec | cheloha | 1 | -5/+5 |
| 2019-01-28 | Simplify by using `spc' since we already have it, no behavior change. | mpi | 1 | -3/+2 |
| 2018-10-17 | Only the scheduler time statistics should be affected by spinning. |   bluhm | 1 | -9/+8 |
| 2018-10-10 | User land time accounting has changed when kernel spinning time was | bluhm | 1 | -3/+5 |
| 2018-06-04 | drop BUMPTIME | cheloha | 1 | -15/+1 |
| 2018-05-14 | Stopping counting and reporting CPU time spent spinning on a lock as | mpi | 1 | -2/+5 |
| 2017-07-22 | Introduce jiffies, a volatile unsigned long version of our ticks variable |   kettenis | 1 | -1/+5 |
| 2017-04-05 | clear structure on the stack before copying out for sysctl. At present |   deraadt | 1 | -1/+2 |
| 2016-09-04 | Introduce Dynamic Profiling, a ddb(4) based & gprof compatible kernel | mpi | 1 | -4/+4 |
| 2016-03-24 | set ticks 15 seconds before its value wraps. |   dlg | 1 | -1/+3 |
| 2016-03-20 | Update ticks in hardclock(). |   uebayasi | 1 | -1/+2 |
| 2015-06-11 | Move hzto(9) to the attic; OK dlg |   mikeb | 1 | -54/+1 |
| 2014-09-15 | Remove non-standard <sys/dkstat.h> header. It has not contained anything |   miod | 1 | -2/+1 |
| 2014-09-04 | Remove global cp_time[] array; no longer used now that all arches implement | miod | 1 | -3/+1 |
| 2014-07-08 | decouple struct uvmexp into a new file, so that uvm_extern.h and sysctl.h | deraadt | 1 | -2/+1 |
| 2013-12-24 | rename local ticks to nticks to avoid aliasing global. ok krw |   tedu | 1 | -17/+17 |
| 2013-10-08 | Fix delivery of SIGPROF and SIGVTALRM to threaded processes by having |   guenther | 1 | -37/+12 |
| 2013-08-13 | Switch time_t, ino_t, clock_t, and struct kevent's ident and data | guenther | 1 | -2/+3 |
| 2013-04-24 | Add tstohz(9) as the timespec analog to tvtohz(9). |   matthew | 1 | -1/+19 |
| 2013-03-28 | do not include machine/cpu.h from a .c file; it is the responsibility of | deraadt | 1 | -2/+1 |
| 2013-03-12 | Fix kernel profiling on MP systems by using per-CPU buffers and teach | mpi | 1 | -3/+3 |
| 2013-02-12 | Back out per-CPU kernel profiling, it shouldn't modify a public header | mpi | 1 | -3/+3 |
| 2013-02-11 | Fix kernel profiling on MP systems by using per-CPU buffer. Previously | mpi | 1 | -3/+3 |
| 2012-11-05 | unifdef -D __HAVE_TIMECOUNTER | miod | 1 | -147/+1 |
| 2012-08-02 | Apply profiling to all threads instead of just the thread that called | guenther | 1 | -10/+12 |
| 2012-05-24 | On resume, run forward the monotonic and realtimes clocks instead of jumping | guenther | 1 | -4/+1 |
| 2012-03-23 | Make rusage totals, itimers, and profile settings per-process instead | guenther | 1 | -17/+12 |
| 2011-03-07 | The scheduling 'nice' value is per-process, not per-thread, so move it | guenther | 1 | -2/+2 |
| 2010-08-20 | Change hzto(9) and tvtohz(9) arguments to const pointers. | matthew | 1 | -3/+3 |
| 2010-01-14 | fix typos in comments, no code changes; |   schwarze | 1 | -2/+2 |
| 2009-11-04 | Get rid of __HAVE_GENERIC_SOFT_INTERRUPTS now that all our platforms support it. | kettenis | 1 | -23/+3 |
| 2009-01-21 | Pass the new pointer to sysctl_clockrate(), so that trying to change the | miod | 1 | -3/+3 |
| 2008-10-04 | At clock initialization, if we discover that tickadj is 0 (due to very | deraadt | 1 | -1/+6 |
| 2008-03-15 | In statclock(), fix local index type in the profiling code. Avoids | miod | 1 | -2/+2 |
| 2007-10-10 | Make context switching much more MI: |   art | 1 | -2/+2 |
| 2007-05-16 | The world of __HAVEs and __HAVE_NOTs is reducing. All architectures | art | 1 | -33/+9 |
| 2007-04-17 | Don't bother performing ITIMER accounting on P_SYSTEM processes (i.e. kthreads). | miod | 1 | -2/+2 |
| 2007-03-25 | Garbage collect CPU_CLOCKUPDATE since it was only used by i386. | art | 1 | -5/+1 |
| 2007-03-15 | Since p_flag is often manipulated in interrupts and without biglock | art | 1 | -3/+3 |
| 2006-12-24 | If <machine/cpu.h> defines PROC_PC to compute the userland pc from an exception | miod | 1 | -1/+6 |
| 2006-06-14 | Introducing adjfreq(2), to adjust the clock frequency. |   otto | 1 | -1/+22 |
| 2006-01-20 | revert auto-skew. some ntpd implementations interact poorly. | tedu | 1 | -3/+2 |
| 2006-01-13 | auto adjusting adjtime. be more responsive to large clock deltas, | tedu | 1 | -2/+3 |
| 2006-01-03 | Make #undef match its temporary #define. | miod | 1 | -2/+2 |
| 2005-11-28 | ansi/deregister. |   jsg | 1 | -14/+9 |
| 2005-10-25 | dont' -> don't in comments; |   jmc | 1 | -2/+2 |
| 2005-10-03 | In hardclock(), do not setup process virtual and profile timers if the process |   aaron | 1 | -2/+2 |
| 2004-11-10 | Diff from art@: |   grange | 1 | -2/+1 |
