aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* Add sched_trace data for Apative EDZLwip-aedzl-revisedGlenn Elliott2010-11-29
| | | | | Added new trace event, st_task_completion_adaptive, to trace estimated execution time and estimation error when an AEDZL task completes.
* Per-task feedback control.Glenn Elliott2010-11-26
| | | | | | | Allow tasks to specify unique feedback control parameters. If no parameters are specified (LITMUS_FP_ZERO), then default parameters of a = .102 and b = .303 are used, a critically- damped feedback control system.
* Merge branch 'wip-edzl-revised' into wip-aedzl-revisedGlenn Elliott2010-11-26
|\
| * Clean up comments.wip-edzl-revisedGlenn Elliott2010-11-26
| | | | | | | | Clean up comments in sched_edzl.c; remove references to GSN-EDF.
* | Implementation of Adaptive-EDZLGlenn Elliott2010-11-26
|/ | | | | | This patch introduces the Adaptive-EDZL (AEDZL) scheduler. AEDZL uses feedback-control to estimate job execution time. This improves the detection of zero-laxity points if WCETs are not tight.
* Implementation of EDZLGlenn Elliott2010-11-26
| | | | | | This patch implements the global EDZL scheduler. It is based heavily on the GSN-EDF implementation. Timers are used to detect zero-laxity points and issue the proper rescheduling operations.
* Litmus core: allow PRECISE_ENFORCEMENT2010.2Bjoern B. Brandenburg2010-10-19
| | | | | Allow all kinds of budget enforcement settings now that we have the supporting infrastructure.
* Litmus core: enable precise budget enforcementBjoern B. Brandenburg2010-10-19
| | | | Update the budget enforcement timer after each scheduling decision.
* Litmus core: add plugin-independent precise budget enforcement infrastructureBjoern B. Brandenburg2010-10-19
| | | | | | | Simple logic: if a task requires precise enforcement, then program a hr-timer to fire when the task must be descheduled. When the timer fires, simply activate the scheduler. When we switch to a different task, either reprogram the timer or cancel it.
* Litmus core: add macro to test for PRECISE_ENFORCEMENTBjoern B. Brandenburg2010-10-19
| | | | | | Required for EDF-WM. We should implement precise enforcement in the core distribution soon anyway (once we know how it works in EDF-WM).
* Litmus core: add budget_remaining() helperBjoern B. Brandenburg2010-10-19
| | | | Quick way to figure out how much budget a LITMUS^RT job has left.
* hrtimer: add init function to properly set hrtimer_start_on_info paramsAndrea Bastoni2010-10-19
| | | | | | | This helper function is also useful to remind us that if we use hrtimer_pull outside the scope of triggering remote releases, we need to take care of properly set the "state" field of hrtimer_start_on_info structure.
* Litmus core: set state to TASK_RUNNING before calling wake_up()Bjoern B. Brandenburg2010-10-19
| | | | | | | | | Having tasks that are !is_running() in shared structures is very confusing during development and debugging, and can likely mask bugs and/or create races. It seems like a strange choice that Linux changes a task's state only _after_ activating it. For LITMUS^RT tasks, we change this order.
* rt_domain_t: disable timer TRACE() spam by defaultBjoern B. Brandenburg2010-10-19
| | | | | | | These messages are highly useful when debugging races, but they quickly litter the log when looking for something else. We keep them around, but by default they shouldn't show up.
* rt_domain_t: add add_release_on()Bjoern B. Brandenburg2010-10-18
| | | | | | | | This API addition allows the calling code to override the release master for a given rt_domain_t object. This is particularly useful if a job is supposed to migrate to a particular CPU. This need arises for example in semi- partitioned schedulers.
* PSN-EDF: remove outdated commentBjoern B. Brandenburg2010-09-21
| | | | | ...and replace it with a more useful one. We don't directly modify Linux run queues anymore since (at least) LITMUS^RT 2008.
* Bugfix: avoid link error in Feather-Trace on x86Bjoern B. Brandenburg2010-07-16
| | | | | | | | | | | | | | | | | | | If no events are defined but Feater-Trace support is enabled, then the current implementation generates a link error because the __event_table sections is absent. > arch/x86/built-in.o: In function `ft_disable_all_events': > (.text+0x242af): undefined reference to `__start___event_table' As a simple work around, we force zero-element array to always be "allocated" in the __event_table section. This ensures that we end up with a zero-byte section if no events are enabled, and does not affect the layout of the section if events are present. > bbb@ludwig:~/dev/litmus2010$ nm vmlinux | grep event_table > ffffffff81950cdc D __event_table_dummy > ffffffff81950cdc A __start___event_table > ffffffff81950cdc A __stop___event_table
* Bugfix: avoid conditional compilation dependent errorAndrea Bastoni2010-07-16
| | | | | | | | | | | | If RELEASE_MASTER is not selected the "info" hrtimer_start_on_info structure in release_heap structure is not visible and trying to access "info" from reinit_release_heap() causes the following error: error: 'struct release_heap' has no member named 'info' info should not be referenced if RELEASE_MASTER is not used. The problem was first reported by Glenn <gelliott@cs.unc.edu>
* Bugfix: change __ARCH_HAS_SEND_PULL_TIMERS in CONFIG_ARCH_HAS_SEND_PULL_TIMERSAndrea Bastoni2010-06-12
| | | | | | | | | Commit "0c527966 Make release master support optional" uses __ARCH_HAS_SEND_PULL_TIMERS instead of CONFIG_ARCH_HAS_SEND_PULL_TIMERS (introduced in commit 0fb33c99) to conditionally compile a pull timer related code in rt_domain.c. This code is disabled and pull-timer's state is no longer properly reset. Therefore, a pulled timer cannot be armed anymore.
* Make litmus_sched_class staticAndrea Bastoni2010-06-10
| | | | | litmus_sched_class wasn't declared static, but it's not used outside sched.c, so change it's signature to static.
* Make platform-specific Feather-Trace depend on !CONFIG_DEBUG_RODATABjoern B. Brandenburg2010-06-01
| | | | | | | | | | | Feather-Trace rewrites instructions in the kernel's .text segment. This segment may be write-protected if CONFIG_DEBUG_RODATA is selected. In this case, fall back to the default flag-based Feather-Trace implementation. In the future, we could either adopt the ftrace method of rewriting .text addresses using non-.text mappings or we could consider replacing Feather-Trace with ftrace altogether. For now, this patch avoids unexpected runtime errors.
* Make PFAIR optional to prevent build and runtime failures.Bjoern B. Brandenburg2010-06-01
| | | | | | | | | | | | | | The PFAIR plugin always implicitly assumed !NO_HZ (the schedule is wrong if NO_HZ is enabled) and does not built if hrtimers are absent: > litmus/built-in.o: In function `pfair_activate_plugin': > sched_pfair.c:(.text+0x7f07): undefined reference to `cpu_stagger_offset' > litmus/built-in.o: In function `init_pfair': > sched_pfair.c:(.init.text+0x487): undefined reference to `cpu_stagger_offset' cpu_stagger_offset() is only available if hrtimers are enabled. This patch makes these dependencies explicit.
* Make C-EDF depend on x86 and SYSFSAndrea Bastoni2010-06-01
| | | | | | C-EDF depends on intel_cacheinfo.c (for get_shared_cpu_map()) which is only available on x86 architectures. Furthermore, get_shared_cpu_map() is only available if SYSFS filesystem is present.
* Make smp_send_pull_timers() optional.Bjoern B. Brandenburg2010-06-01
| | | | | There is currently no need to implement this in ARM. So let's make it optional instead.
* Make release master support optionalBjoern B. Brandenburg2010-06-01
| | | | | | | | Introduces CONFIG_RELEASE_MASTER and makes release master support dependent on the new symbol. This is useful because dedicated interrupt handling only applies to "large" multicore platforms. This will allow us to not implement smp_send_pull_timers() for all platforms.
* Make compilation of C-EDF optional.Bjoern B. Brandenburg2010-05-30
| | | | | | C-EDF only makes sense on multicore platforms that have shared caches. Make it possible to disable it on other platforms, in particular, on those that do not export get_shared_cpu_map().
* Make __ARCH_HAS_FEATHER_TRACE a proper CONFIG_ variable.Bjoern B. Brandenburg2010-05-30
| | | | | | | | | The idea of the Feather-Trace default implementation is that LITMUS^RT should work without a specialized Feather-Trace implementation present. This was actually broken. Changes litmus/feather_trace.h to only include asm/feather_trace.h if actually promised by the architecture.
* Bugfix: re-insert missing TS_PLUGIN_TICK_END tracing pointAndrea Bastoni2010-05-30
| | | | | Insert PLUGIN_TICK_END tracing point in litmus_tick(). It was lost during the porting of 2008.3 to 2010.1.
* Add support for one single cluster (all cpus) on C-EDFAndrea Bastoni2010-05-30
| | | | | - With the "ALL" cluster size option the behavior of C-EDF is equivalent to G-EDF (one single cluster)
* Added support for choices in budget policy enforcement.Glenn Elliott2010-05-30
| | | | | | | | | | NO_ENFORCEMENT - A job may execute beyond its declared execution time. Jobs notify the kernel that they are complete via liblitmus's sleep_next_period() QUANTUM_ENFORCEMENT - The kernel terminates a job if its actual execution time exceeds the declared execution time. PRECISE_ENFORCEMENT - Hook declared, but not yet implemented. Plan to support this policy through hrtimers. Error thrown if specified.
* Update kfifo and spinlock_t in sched_trace.cAndrea Bastoni2010-05-29
| | | | | | | | | - kfifo needs to be defined and used differently (see include/linux/kfifo.h) - spinlock -> raw_spinlock - include slab.h when using kmalloc and friends This commit compiles and is the logical end of the merge of Litmus and 2.6.34.
* Update sched_class and spinlock_t in litmus.cAndrea Bastoni2010-05-29
| | | | | | | | - get_rr_interval() changed signature - load_balance() and move_one_tak() are no longer needed - spinlock_t -> raw_spinlock_t This commit does not compile.
* Change most LitmusRT spinlock_t in raw_spinlock_tAndrea Bastoni2010-05-29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Adapt to new schema for spinlock: (tglx 20091217) spinlock - the weakest one, which might sleep in RT raw_spinlock - spinlock which always spins even on RT arch_spinlock - the hardware level architecture dependent implementation ---- Most probably, all the spinlocks changed by this commit will be true spinning lock (raw_spinlock) in PreemptRT (so hopefully we'll need few changes when porting Litmmus to PreemptRT). There are a couple of spinlock that the kernel still defines as spinlock_t (therefore no changes reported in this commit) that might cause us troubles: - wait_queue_t lock is defined as spinlock_t; it is used in: * fmlp.c -- sem->wait.lock * sync.c -- ts_release.wait.lock - rwlock_t used in fifo implementation in sched_trace.c * this need probably to be changed to something always spinning in RT at the expense of increased locking time. ---- This commit also fixes warnings and errors due to the need to include slab.h when using kmalloc() and friends. ---- This commit does not compile.
* Merge branch 'master' into wip-merge-2.6.34Andrea Bastoni2010-05-29
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Simple merge between master and 2.6.34 with conflicts resolved. This commit does not compile, the following main problems are still unresolved: - spinlock -> raw_spinlock API changes - kfifo API changes - sched_class API changes Conflicts: Makefile arch/x86/include/asm/hw_irq.h arch/x86/include/asm/unistd_32.h arch/x86/kernel/syscall_table_32.S include/linux/hrtimer.h kernel/sched.c kernel/sched_fair.c
| * Add C-EDF Plugin2010.1Andrea Bastoni2010-05-29
| | | | | | | | | | | | Improved C-EDF plugin. C-EDF now supports different cluster sizes (based on L2 and L3 cache sharing) and supports dynamic changes of cluster size (this requires reloading the plugin).
| * Export shared_cpu_mapAndrea Bastoni2010-05-29
| | | | | | | | | | | | The cpumap of CPUs that share the same cache level is not normally available outside intel_cacheinfo.c. This commit allows to export such map.
| * Synchronize plugin switchingAndrea Bastoni2010-05-29
| | | | | | | | | | | | Make sure the plugin is not used by any CPUs while switching. The CPU performing the switch sends an IPI to all other CPUs forcing them to synchronize on an atomic variable.
| * Measure timer re-arming in the proper locationAndrea Bastoni2010-05-29
| | | | | | | | | | | | hrtimers are properly rearmed during arm_release_timer() and no longer after rescheduling (with the norqlock mechanism of 2008.3). This commit accordingly updates the locations where measures are taken.
| * Bugfix: PSN-EDF should log job_completion eventsAndrea Bastoni2010-05-29
| | | | | | | | | | Log task completions in job_completion() for PSN-EDF. This fixes the problem of missing job-completion events for PSN-EDF.
| * Bugfix: PSN-EDF should only requeue tasks that are not scheduledBjoern B. Brandenburg2010-05-29
| | | | | | | | | | | | | | | | | | Requeue a task that is already scheduled will cause it to be effectively in the runqueue twice since scheduled tasks are conceptually the head of the queue. If a task is still scheduled, then schedule() will do the right thing and do the requeuing if necessary. This fixes crashes reported by Glenn and Andrea.
| * Used miscdevice API for sched_traceBjoern B. Brandenburg2010-05-29
| | | | | | | | | | | | | | | | | | This patch changes sched_trace.c to use the miscdevice API instead of doing all the cdev management ourselves. This remove a chunk of code and we get sysfs / udev integration for free. On systems with default udev rules, this will result in a /dev/litmus/log device being created automatically.
| * Bugfix: make fdso syscalls 64bit cleanBjoern B. Brandenburg2010-05-29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This fixes a bug found by liblitmus's regression test suite. Before: > ** LITMUS^RT test suite. > ** Running tests for LINUX. > ** Testing: don't open FMLP semaphores if FMLP is not supported... > !! TEST FAILURE open_fmlp_sem(fd, 0) -> -16, Success (expected: EBUSY) > at tests/fdso.c:21 (test_fmlp_not_active) > ** Testing: reject invalid object descriptors... ok. > ** Testing: reject invalid object types... > !! TEST FAILURE od_open(0, -1, 0) -> -22, Bad file descriptor (expected: EINVAL) > at tests/fdso.c:51 (test_invalid_obj_type) > ** Testing: reject invalid rt_task pointers... ok. > ** Result: 2 ok, 2 failed. After: > ** LITMUS^RT test suite. > ** Running tests for LINUX. > ** Testing: don't open FMLP semaphores if FMLP is not supported... ok. > ** Testing: reject invalid object descriptors... ok. > ** Testing: reject invalid object types... ok. > ** Testing: reject invalid rt_task pointers... ok. > ** Result: 4 ok, 0 failed.
| * Bugfix: don't inherit od_table across forksBjoern B. Brandenburg2010-05-29
| | | | | | | | | | | | | | | | The od_table is strictly per-thread and should not be inherited across a fork/clone. This caused memory corruption when a task exited, which ultimately could lead to oopses in unrelated code. Bug and testcase initially reported by Glenn.
| * Bugfix: 1) incorrect FMLP high prio task tracking and 2) race in print statementGlenn Elliott2010-05-29
| | | | | | | | | | | | | | | | 1) High priority task tied to FMLP semaphore in P-EDF scheduling is incorrectly tracked for tasks acquiring the lock without contention. (HP is always set to CPU 0 instead of proper CPU.) 2) Race in a print statement from P-EDF's pi_block() causes NULL pointer dereference.
| * Use generic preemption function in GSN- and PSN-EDF.Bjoern B. Brandenburg2010-05-29
| | | | | | | | | | This patch updates non-preemptive section support in GSN- and PSN-EDF.
| * Introduce generic NP-section aware preemption functionBjoern B. Brandenburg2010-05-29
| | | | | | | | | | | | | | | | Dealing with preemptions across CPUs in the presence of non-preemptive sections can be tricky and should not be replicated across (event-driven) plugins. This patch introduces a generic preemption function that handles non-preemptive sections (hopefully) correctly.
| * Re-implement non-preemptive section support.Bjoern B. Brandenburg2010-05-29
| | | | | | | | | | | | | | | | | | | | | | | | Re-introduce NP sections in the configuration and in litmus.h. Remove the old np_flag from rt_param. If CONFIG_NP_SECTION is disabled, then all non-preemptive section checks are constant expressions which should get removed by the dead code elimination during optimization. Instead of re-implementing sys_exit_np(), we simply repurposed sched_yield() for calling into the scheduler to trigger delayed preemptions.
| * Add virtual LITMUS^RT control device.Bjoern B. Brandenburg2010-05-29
| | | | | | | | | | | | | | | | | | This device only supports mmap()'ing a single page. This page is shared RW between the kernel and userspace. It is inteded to allow near-zero-overhead communication between the kernel and userspace. It's first use will be a proper implementation of user-signaled non-preemptable section support.
| * Bugfix: clear LITMUS^RT state on fork completelyBjoern B. Brandenburg2010-05-29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When a real-time task forks, then its LITMUS^RT-specific fields should be cleared, because we don't want real-time tasks to spawn new real-time tasks that bypass the plugin's admission control (if any). This was broken in three ways: 1) kernel/fork.c did not erase all of tsk->rt_param, only the first few bytes due to a wrong size argument to memset(). 2) It should have been calling litmus_fork() instead anyway. 3) litmus_fork() was _also_ not clearing all of tsk->rt_param, due to another size argument bug. Interestingly, 1) and 2) can be traced back to the 2007->2008 port, whereas 3) was added by Mitchell much later on (to dead code, no less). I'm really surprised that this never blew up before.
| * Add Feather-Trace x86_64 architecture dependent codeAndrea Bastoni2010-05-29
| |
generated by cgit v1.2.2 (git 2.25.0) at 2024-11-21 04:52:44 -0500