| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | There was a thread on "unusually high load averages" when running under | obrien | 2004-04-22 | 1 | -2/+2 |
| * | Spell "switches" a more conventional way. | cognet | 2004-04-09 | 1 | -1/+1 |
| * | - Use the proper constant in sched_interact_update(). Previously, | jeff | 2004-04-04 | 1 | -1/+1 |
| * | Change the type of the various CPU masks to cpumask_t. Note that as | marcel | 2004-03-27 | 1 | -4/+4 |
| * | Give a more reasonable CPU time to the threads which are using scheduler | obrien | 2004-03-21 | 1 | -6/+3 |
| * | Switch the sleep/wakeup and condition variable implementations to use the | jhb | 2004-02-27 | 1 | -2/+2 |
| * | - Allow interactive tasks to use the maximum time-slice. This is not as | jeff | 2004-02-01 | 1 | -1/+1 |
| * | - Add a new member to struct kseq called ksq_sysload. This is intended to | jeff | 2004-02-01 | 1 | -3/+27 |
| * | - sched_strict has been dead for a long time now. Get rid of it. | jeff | 2004-01-25 | 1 | -3/+0 |
| * | - Clean up KASSERTS. | jeff | 2004-01-25 | 1 | -4/+4 |
| * | - Add a flags parameter to mi_switch. The value of flags may be SW_VOL or | jeff | 2004-01-25 | 1 | -2/+1 |
| * | - Make our transfer decisions based on load and not transferable load. A | jeff | 2003-12-20 | 1 | -7/+1 |
| * | - Enable ithread migration on x86. This is done to work around a bug in the | jeff | 2003-12-20 | 1 | -0/+10 |
| * | - In kseq_transfer() return if smp has not been started. | jeff | 2003-12-20 | 1 | -9/+14 |
| * | - Running interactive tasks with the minimum time-slice is fine for vi and | jeff | 2003-12-20 | 1 | -1/+2 |
| * | - Assign the ke_cpu field in kseq_notify() so that all of our callers do not | jeff | 2003-12-14 | 1 | -4/+2 |
| * | - Now that we have kseq groups, balance them seperately. | jeff | 2003-12-12 | 1 | -47/+130 |
| * | - Don't let the pctcpu rate limiter throttle us if we have recorded over | jeff | 2003-12-11 | 1 | -1/+2 |
| * | - In sched_switch(), if a thread has been assigned, don't touch the runqueues | jeff | 2003-12-11 | 1 | -15/+21 |
| * | - Add support for CPU groups to ule. All SMT cores on the same physical | jeff | 2003-12-11 | 1 | -116/+263 |
| * | rqb_bits[] may be an int64_t (eg: on alpha, and recently on amd64). | peter | 2003-12-07 | 1 | -1/+1 |
| * | Fix all users of mp_maxid to use the same semantics, namely: | jhb | 2003-12-03 | 1 | -1/+1 |
| * | - Mark ksq_assigned as volatile so that when this code is used without | jeff | 2003-11-17 | 1 | -3/+3 |
| * | - Remove long dead code. rslices hasn't been used in some time and neither | jeff | 2003-11-17 | 1 | -52/+4 |
| * | - Introduce kseq_runq_{add,rem}() which are used to insert and remove | jeff | 2003-11-15 | 1 | -61/+83 |
| * | - Somehow I botched my last commit. Add an extra ( to fix things up. I'm | jeff | 2003-11-06 | 1 | -1/+1 |
| * | - Remove the local definition of sched_pin and unpin. They are provided in | jeff | 2003-11-06 | 1 | -17/+3 |
| * | - It's ok if sched_runnable() has races in it, we don't need the sched_lock | jeff | 2003-11-05 | 1 | -3/+4 |
| * | - Add initial support for pinning and binding. | jeff | 2003-11-04 | 1 | -2/+53 |
| * | - Remove kseq_find(), we no longer scan other cpu's run queues when we go | jeff | 2003-11-03 | 1 | -66/+17 |
| * | - Remove the ksq_loads[] array. We are only interested in three counts, | jeff | 2003-11-02 | 1 | -33/+50 |
| * | - In sched_prio() only force us onto the current queue if our priority is | jeff | 2003-11-02 | 1 | -1/+2 |
| * | - Rename SCHED_PRI_NTHRESH to SCHED_SLICE_NTHRESH since it is only used in | jeff | 2003-11-02 | 1 | -10/+11 |
| * | - Remove uses of PRIO_TOTAL and replace them with SCHED_PRI_NRESV | jeff | 2003-11-02 | 1 | -5/+5 |
| * | - Change sched_interact_update() to only accept slp+runtime values between | jeff | 2003-11-02 | 1 | -27/+56 |
| * | - Add static to local functions and data where it was missing. | jeff | 2003-10-31 | 1 | -78/+222 |
| * | Removed sched_nest variable in sched_switch(). Context switches always | bde | 2003-10-29 | 1 | -3/+0 |
| * | - Only change the run queue in sched_prio() if the kse is non null. threads | jeff | 2003-10-28 | 1 | -10/+2 |
| * | - Use a better algorithm in sched_pctcpu_update() | jeff | 2003-10-27 | 1 | -56/+50 |
| * | - If a thread is not bound to a kse return 0 from sched_pctcpu(). | jeff | 2003-10-20 | 1 | -0/+2 |
| * | - Only kse_reassign() in the !running case. | jeff | 2003-10-16 | 1 | -8/+10 |
| * | - Call sched_add() with the correct argument on SMP. | jeff | 2003-10-16 | 1 | -1/+1 |
| * | - Fix a minor problem with my last commit, we don't want to return from | jeff | 2003-10-16 | 1 | -3/+1 |
| * | - Collapse sched_switchin() and sched_switchout() into sched_switch(). Now | jeff | 2003-10-16 | 1 | -8/+9 |
| * | - Update the sched api. sched_{add,rem,clock,pctcpu} now all accept a td | jeff | 2003-10-16 | 1 | -7/+14 |
| * | - The non iterative algorithm for interact_update was broken due to | jeff | 2003-10-16 | 1 | -8/+6 |
| * | - If our user_pri doesn't match our actual priority our priority has been | jeff | 2003-10-15 | 1 | -3/+10 |
| * | - In SCHED_CURR() add holding Giant to the list of criteria that will keep | jeff | 2003-10-12 | 1 | -8/+7 |
| * | - Change a lame iterative algorithm to a constant time algorithm. Remove | jeff | 2003-10-04 | 1 | -4/+6 |
| * | - Somewhere along the line I stupidly removed critical logic from | jeff | 2003-09-20 | 1 | -10/+11 |
