aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* Description of refined aux task inheritance.wip-aux-tasksGlenn Elliott2012-09-17
|
* Fixed three bugs with aux threads and nested locksGlenn Elliott2012-09-17
| | | | | | | | | | | | | | Fixes two bugs with nested locks: 1) List of aux threads could become corrupted. -- moved modifications to be within scheduler lock. 2) Fixed bad EDF comparison ordering that could lead to schedule thrashing in an infinite loop. 3) Prevent aux threads from inheriting a priority from a task that is blocked on a real-time litmus lock. (since the aux threads can't possibly hold these locks, we don't have to worry about inheritance.)
* Merge branch 'prop/sched_fork' into wip-aux-tasksGlenn Elliott2012-09-17
|\ | | | | | | | | Conflicts: kernel/sched.c
| * Do processor state transitions in schedule_tail().Glenn Elliott2012-09-17
| | | | | | | | | | | | | | | | | | Fixes a bug in Litmus where processor scheduling states could become corrupted. Corruption can occur when a just-forked thread is externally forced to be scheduled by SCHED_LITMUS before this just-forked thread can complete post-fork processing. Specifically, before schedule_tail() has completed.
* | C-EDF support for auxillary tasks.Glenn Elliott2012-09-16
| | | | | | | | Extended auxillary task support to C-EDF. Modeld after G-EDF.
* | Added CONFIG_REALTIME_AUX_TASKS optionGlenn Elliott2012-09-16
| | | | | | | | | | | | Auxillary task features were enabled by CONFIG_LITMUS_LOCKING. Made auxillary tasks a seperate feature that depends upon CONFIG_LITMUS_LOCKING.
* | Implement real-time aux threads. G-EDF only.Glenn Elliott2012-09-16
| |
* | checkpoint for aux_tasks. can still deadlockGlenn Elliott2012-09-14
| |
* | Merge branch 'wip-gpu-rtas12' into wip-slave-threadsGlenn Elliott2012-09-11
|\ \ | | | | | | | | | | | | | | | | | | Conflicts: include/litmus/unistd_32.h include/litmus/unistd_64.h litmus/litmus.c
| * | changed gpu filtering to 1.5 stdev.Glenn Elliott2012-09-11
| | |
| * | Fix hang from bug in edf_common.cGlenn Elliott2012-09-11
| | |
| * | Merge remote-tracking branch 'github/prop/litmus-signals' into wip-gpu-rtas12Glenn Elliott2012-09-10
| |\ \ | | | | | | | | | | | | | | | | Conflicts: litmus/sched_gsn_edf.c
| | * | Infrastructure for Litmus signals.prop/litmus-signalsGlenn Elliott2012-09-07
| | |/ | | | | | | | | | | | | | | | | | | | | | Added signals to Litmus. Specifcally, SIG_BUDGET signals are delivered (when requested by real-time tasks) when a budget is exceeded. Note: pfair not currently supported (but it probably could be).
| * | Merge branch 'prop/robust-tie-break' into wip-gpu-rtas12Glenn Elliott2012-09-10
| |\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Conflicts: include/litmus/binheap.h include/litmus/fdso.h include/litmus/litmus.h litmus/Makefile litmus/binheap.c litmus/edf_common.c litmus/fdso.c litmus/jobs.c litmus/locking.c
| | * | Fixed type-os and clarified text in litmus/KconfigGlenn Elliott2012-09-08
| | | |
| | * | EDF priority tie-breaks.wip-robust-tie-breakGlenn Elliott2012-08-27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of tie-breaking by PID (which is a static priority tie-break), we can tie-break by other job-level-unique parameters. This is desirable because tasks are equaly affected by tardiness since static priority tie-breaks cause tasks with greater PID values to experience the most tardiness. There are four tie-break methods: 1) Lateness. If two jobs, J_{1,i} and J_{2,j} of tasks T_1 and T_2, respectively, have equal deadlines, we favor the job of the task that had the worst lateness for jobs J_{1,i-1} and J_{2,j-1}. Note: Unlike tardiness, lateness may be less than zero. This occurs when a job finishes before its deadline. 2) Normalized Lateness. The same as #1, except lateness is first normalized by each task's relative deadline. This prevents tasks with short relative deadlines and small execution requirements from always losing tie-breaks. 3) Hash. The job tuple (PID, Job#) is used to generate a hash. Hash values are then compared. A job has ~50% chance of winning a tie-break with respect to another job. Note: Emperical testing shows that some jobs can have +/- ~1.5% advantage in tie-breaks. Linux's built-in hash function is not totally a uniform hash. 4) PIDs. PID-based tie-break used in prior versions of Litmus. Conflicts: litmus/edf_common.c
| | * | Improve readability of EDF comparisons.Glenn Elliott2012-08-20
| | |/ | | | | | | | | | | | | | | | Restructured the EDF task comparison code to improve readability. Recoded chained logical expression embedded in return statement into a series of if/else blocks.
| * | standard devation-based gpu affinity predictorGlenn Elliott2012-09-10
| | |
| * | simple average trackingGlenn Elliott2012-09-09
| | |
| * | enable migration tracing and short-cut interruptsGlenn Elliott2012-08-31
| | |
| * | GPUSync patch for Litmus 2012.1.Glenn Elliott2012-05-26
| | |
* | | blargGlenn Elliott2012-09-11
| |/ |/|
* | New Feature: Arbitrary deadlines.2012.2Glenn Elliott2012-08-01
| | | | | | | | | | | | | | | | | | | | | | | | Added support for arbitrary deadlines. Constraint: Relative deadline must be >= exec cost. Use: Set relative deadline in rt_task::rdeadline. Set value to 0 to default to implicit deadlines. Limitations: PFAIR not supported by this patch. PFAIR updated to reject tasks that do not have implicit deadlines.
* | control page: avoid "minor" page faultsBjoern Brandenburg2012-07-23
| | | | | | | | | | | | | | | | | | By default, even private writable pages are mapped with the RW bit disabled in the PTE. This causes a "minor" page fault when the page is first written to. To avoid this, make sure that vm_inert_page() uses the proper page protection bits and mark the VMA as VM_IO to keep the rest of the VM code out.
* | control page: warn on page faultBjoern Brandenburg2012-07-23
| | | | | | | | | | Page faults should not happen here. Scream if they do anyway. This is useful when extending the control page.
* | control page: use vm_insert_page() instead of remap_pfn_range()Bjoern Brandenburg2012-07-23
| | | | | | | | | | | | vm_insert_page() is the simpler and preferred interface for remapping individual pages and includes additional error checks. It suffices for our purposes, so let's use it instead.
* | P-FP: reject tasks with invalid prioritiesBjoern Brandenburg2012-07-23
| | | | | | | | | | | | The existing admission test failed to test for too-low priorities. Use the common macro to accept only valid priorities.
* | P-FP: Clarify meaning of prioritiesBjoern Brandenburg2012-07-23
| | | | | | | | | | | | Add a comment to explain how priorities are interpreted, and provide some useful macros for userspace.
* | P-FP: fix build bug if !CONFIG_RELEASE_MASTERBjoern Brandenburg2012-07-23
| | | | | | | | | | Move declaration of 'cpu' out of #ifdef block, it's also needed for CONFIG_LITMUS_LOCKING.
* | P-FP: make PCP available to userspaceSven Dziadek2012-07-23
| | | | | | | | Prior to that it was only used internally for DPCP
* | P-FP: port P-FP plugin used in B. Brandenburg'sSven Dziadek2012-07-23
| | | | | | | | | | | | | | | | | | | | dissertation (branch bbb-diss) to current version of litmus This is needed for ongoing projects I took the unchanged code but removed some leftovers of OMLP which is not implemented
* | Update to improved x86 Feather-Trace triggersBjoern Brandenburg2012-06-26
| | | | | | | | | | | | | | | | | | | | This patch imports recent upstream changes in Feather-Trace that reduce register pressure around Feather-Trace triggers. References: Commits 00713b8 and 225d734 in Feather-Trace. https://github.com/brandenburg/feather-trace/commit/00713b878636867ce07291c588509b38fa5bf152 https://github.com/brandenburg/feather-trace/commit/225d7348a08682cd87f72b127142bdfd6c0c7890
* | C-EDF: do not requeue jobs without budgetBjoern Brandenburg2012-05-31
| | | | | | | | | | | | This patch replicates the fix in commit f141d730e91283a9bb5cfcb134fcead55d5da0c6 (which applies to GSN-EDF).
* | GSN-EDF: do not requeue jobs without budgetBjoern Brandenburg2012-05-31
| | | | | | | | | | | | | | | | | | | | This patch changes how preemptions of jobs without budget work. Instead of requeuing them, they are now only added if they are not subject to budget enforcement or if they have non-zero budget. This allows us to process job completions that race with preemptions. This appears to fix a BUG in budget.c:65 reported by Giovani Gracioli.
* | Move budget-related helpers to budget.hBjoern Brandenburg2012-05-31
| | | | | | | | | | | | | | | | | | litmus.h is accumulating too many things. Since we already have budget.h, let's stick all budget-related inline functions there as well. This patch is merely cosmetic; it does not change how budget enforcement works.
* | An efficient binary heap implementation.wip-stage-binheapGlenn Elliott2012-05-26
| | | | | | | | | | | | | | An efficient binary heap implementation coded in the style of Linux's list. This binary heap should be able to replace any partially sorted priority queue based upon Linux's list.
* | Add kernel-style events for sched_trace_XXX() functionsAndrea Bastoni2012-03-30
|/ | | | | | | | | | | | | | | | | | Enable kernel-style events (tracepoint) for Litmus. Litmus events trace the same functions as the sched_trace_XXX(), but can be enabled independently. So, why another tracing infrastructure then: - Litmus tracepoints can be recorded and analyzed together (single time reference) with all other kernel tracing events (e.g., sched:sched_switch, etc.). It's easier to correlate the effects of kernel events on litmus tasks. - It enables a quick way to visualize and process schedule traces using trace-cmd utility and kernelshark visualizer. Kernelshark lacks unit-trace's schedule-correctness checks, but it enables a fast view of schedule traces and it has several filtering options (for all kernel events, not only Litmus').
* Typo in macroJonathan Herman2012-02-16
|
* Feather-Trace: keep track of interrupt-related interference.2012.1Bjoern B. Brandenburg2012-01-30
| | | | | | | Increment a processor-local counter whenever an interrupt is handled. This allows Feather-Trace to include a (truncated) counter and a flag to report interference from interrupts. This could be used to filter samples that were disturbed by interrupts.
* Litmus core: simplify np-section protocolwip-2011.2-bbbBjoern B. Brandenburg2011-11-24
| | | | | | User a 32-bit word for all non-preemptive section flags. Set the "please yield soon" flag atomically when accessing it on remotely-scheduled tasks.
* C-EDF: rename lock -> cluster_lockBjoern B. Brandenburg2011-11-24
| | | | The macro lock conflicts with locking protocols...
* locking: use correct timestampBjoern B. Brandenburg2011-11-24
|
* Feather-trace: let userspace add overhead eventsBjoern B. Brandenburg2011-11-24
| | | | | This is useful for measuring locking-related overheads that are partially recorded in userspace.
* ftdev: let bufffer-specific code handle writes from userspaceBjoern B. Brandenburg2011-11-24
| | | | | This allows us to splice in information into logs from events that were recorded in userspace.
* ftdev: remove event activation hackBjoern B. Brandenburg2011-11-24
| | | | | Instead of doing the hackisch 'write commands to device' thing, let's just use a real ioctl() interface.
* Feather-Trace: keep track of release latencyBjoern B. Brandenburg2011-11-24
|
* Feather-Trace: trace locking-related suspensionsBjoern B. Brandenburg2011-11-24
|
* Feather-Trace: start with the largest permissible rangeBjoern B. Brandenburg2011-11-24
| | | | MAX_ORDER is 11, but this is about number of records, not number of pages.
* bugfix: add processors in order of increasing indices to clustersBjoern B. Brandenburg2011-11-24
| | | | | Pfair expects to look at processors in order of increasing index. Without this patch, Pfair could deadlock in certain situations.
* Pfair: improve robustness of suspensionsBjoern B. Brandenburg2011-11-24
| | | | | | | | | | | | | | | This patch fixes two crash or hang bugs related to suspensions in Pfair. 1) When a job was not present at the end of its last subtask, then its linked_on field was not cleared. This confused the scheduler when it later resumed. Fix: clear the field. 2) Just testing for linked_on == NO_CPU is insufficient in the wake_up path to determine whether a task should be added to the ready queue. If the task remained linked and then was "preempted" at a later quantum boundary, then it already is in the ready queue and nothing is required. Fix: encode need to requeue in task_rt(t)->flags.
generated by cgit v1.2.2 (git 2.25.0) at 2024-11-21 05:32:34 -0500