summaryrefslogtreecommitdiffstats
path: root/testsuites/smptests (unfollow)
Commit message (Collapse)AuthorFilesLines
2019-12-19config: Improve EDF SMP scheduler configurationSebastian Huber5-7/+7
Use CONFIGURE_MAXIMUM_PROCESSORS to configure the EDF SMP scheduler context. This avoids hard to debug configuration errors resulting in memory corruptions. Close #3815.
2019-12-11rtems: Simplify semaphore configurationSebastian Huber3-3/+0
The MrsP semaphore implementation predates the addition of self-contained synchronization objects. At this time, the potential memory reduction was justified considering the more complex configuration and additional use of the workspace. With the availability of self-contained synchronization options, e.g. POSIX mutexes, this is no longer justified. Memory constrained applications should use the self-contained synchronization objects. Remove the CONFIGURE_MAXIMUM_MRSP_SEMAPHORES configuration option. This has only an impact on applications which use SMP and a large number of scheduler instances. Update #3833.
2019-12-05libtest: Change expected pass state stringSebastian Huber5-5/+5
Use separator character '_' for all test states.
2019-11-12tests: Remove superfluous SMPTESTS defineSebastian Huber1-7/+7
Update #3818.
2019-05-20score: Add _SMP_Unicast_action()Sebastian Huber2-43/+151
2019-05-20score: Add _SMP_Synchronize()Sebastian Huber1-3/+4
2019-05-20smpipi01: Use per-CPU jobs for IPI flood testSebastian Huber1-9/+38
2019-05-20smpipi01: Ensure IPI works on all processorsSebastian Huber1-8/+21
2019-05-20smpipi01: Use per-CPU jobs for pending IPI testSebastian Huber1-15/+35
2019-05-20score: Modify _Per_CPU_Perform_jobs()Sebastian Huber1-0/+41
Process only the jobs initially registered on the processing list. This makes it possible to add jobs for the current processor in a job handler. These jobs are processed with the next SMP_MESSAGE_PERFORM_JOBS message. The lock is only acquired and released once.
2019-05-20score: Add _Per_CPU_Add_job()Sebastian Huber1-0/+43
2019-05-20score: Simplify _SMP_Multicast_action()Sebastian Huber1-10/+4
Move resposibility to disable thread dispatching to the caller of _SMP_Multicast_action(). Using an interrupt disable for this purpose is questionable.
2019-05-20score: Add _SMP_Broadcast_action()Sebastian Huber1-8/+32
2019-05-20smptests: Move SMP broadcast action test caseSebastian Huber2-51/+79
2019-05-20smpmulticast01: Use T_TEST_CASE()Sebastian Huber1-22/+27
2019-04-12score: More robust _SMP_Multicast_action()Sebastian Huber2-7/+1
If the caller already disabled interrupts, then do not disable thread dispatching. Calling _SMP_Multicast_action() with interrupts disabled is a questionable use case.
2019-04-12score: Improve _SMP_Multicast_action()Sebastian Huber2-73/+127
Let it work during system initialization.
2019-04-12score: Rework SMP multicast actionSebastian Huber2-1/+7
Use a FIFO list of jobs per processor to carry out the SMP multicast action. Use a done indicator per job to reduce the bus traffic a bit.
2019-04-12score: Use processor mask in _SMP_Multicast_actionSebastian Huber6-38/+386
Processor_mask is the internal data type to deal with processor sets.
2019-04-09rtems: Add rtems_scheduler_get_processor_maximum()Sebastian Huber43-71/+86
Add rtems_scheduler_get_processor_maximum() as a replacement for rtems_get_processor_count(). The rtems_get_processor_count() is a bit orphaned. Adopt it by the Scheduler Manager. The count is also misleading, since the processor set may have gaps and the actual count of online processors may be less than the value returned by rtems_get_processor_count(). Update #3732.
2019-04-09rtems: Add rtems_scheduler_get_processor()Sebastian Huber31-115/+115
Add rtems_scheduler_get_processor() as a replacement for rtems_get_current_processor(). The rtems_get_current_processor() is a bit orphaned. Adopt it by the Scheduler Manager. This is in line with the glibc sched_getcpu() function. Deprecate rtems_get_current_processor(). Update #3731.
2019-02-07score: Fix plain priority thread queues (SMP)Sebastian Huber1-0/+63
We must add/remove the priority queue to the FIFO of priority queues.
2019-01-30build: Do not install test programsSebastian Huber1-3/+1
2019-01-09rtems: Allow to set ISR level 0 in SMP configSebastian Huber1-1/+1
Update #3000.
2018-12-18score: Fix per-CPU data allocationSebastian Huber5-0/+117
Allocate the per-CPU data for secondary processors directly from the heap areas before heap initialization and not via _Workspace_Allocate_aligned(). This avoids dependency on the workspace allocator. It fixes also a problem on some platforms (e.g. QorIQ) where at this early point in the system initialization the top of the RAM is used by low-level startup code on secondary processors (boot pages). Update #3507.
2018-12-04Spelling and grammar fixes in source code comments (GCI 2018)Marçal Comajoan Cara1-1/+1
2018-11-26Remove rtems_cache_*_processor_set() functionsSebastian Huber1-45/+7
The following rtems_cache_*_processor_set() cache manager API functions are exotic, complex, very hard to use correctly, not used in the RTEMS code base, and apparently unused by applications. Close #3622.
2018-10-29posix: Enable more smptests tests by defaultSebastian Huber1-6/+0
Update #2514.
2018-10-10build: Remove local.amSebastian Huber1-1/+0
2018-10-05tests: Use rtems_task_exit()Sebastian Huber4-6/+4
Update #3533.
2018-10-05score: Remove CPU_PROVIDES_IDLE_THREAD_BODYSebastian Huber2-12/+0
Remove the CPU_PROVIDES_IDLE_THREAD_BODY option to avoid unnecessary conditional compilation. Close #3539.
2018-09-20smp03: make printout match actual task nameDaniel Hellstrom1-1/+1
2018-09-10score: Add thread pin/unpin supportSebastian Huber5-0/+679
Add support to temporarily pin a thread to its current processor. This may be used to access per-processor data structures in critical sections with enabled thread dispatching, e.g. a pinned thread is allowed to block. Update #3508.
2018-09-03score: Fix EDF SMP schedulerSebastian Huber1-0/+24
Fix a special case: block a one-to-one scheduled thread while having a non-empty affine ready queue on the same processor.
2018-07-25bsps: bsp_start_on_secondary_processor()Sebastian Huber1-2/+4
Pass current processor control as first parameter in bsp_start_on_secondary_processor() and qoriq_start_thread() to make dependency more explicit.
2018-07-24smptests: Fix format warningsSebastian Huber2-14/+36
2018-07-18score: Fix _Scheduler_EDF_SMP_Set_affinity()Sebastian Huber5-0/+172
Commit 8744498752ad4f0eaf9fb5640c6a0e0f2dc92fda broke the _Scheduler_EDF_SMP_Set_affinity() implementation. We must test the overall affinity against the online processors.
2018-06-28riscv: Add dummy SMP supportSebastian Huber1-1/+2
Update #3433.
2018-05-02tests: Remove configure feature checksSebastian Huber3-50/+1
Update #3409.
2018-04-27smptests/smpschededf02: Add test caseSebastian Huber1-3/+8
2018-04-23smptests/smpschedaffinity02: Fix configurationSebastian Huber1-0/+2
2018-04-11configure: Add subdir-objects to all automake flags.Chris Johns1-1/+1
This option silences warning with automake-1.16.1 allowing us to upgrade to that version. This change has been tested with automake-1.12.6 and automake-1.16.1. It seems version 1.16.1 configures slower than 1.12.6 for the same source and BSP. The newer versions is 6 second slower. Close #3387.
2018-04-10testsuite/smptests: Merged nested Makefile.am files into one Makefile.amChris Johns57-1191/+691
This change is part of the testsuite Makefile.am reorganization. Update #3382
2018-03-12config: Use new scheduler configuration definesSebastian Huber17-304/+304
Update #3325.
2018-03-07smptests: Fix obsolete FIXME comments/codeSebastian Huber2-4/+2
2018-02-02smptests/smpopenmp01: New testSebastian Huber7-0/+510
2018-02-02smpschededf01: Use rtems_test_busy_cpu_usage()Sebastian Huber1-11/+5
The rtems_test_busy() is not accurate enough and may lead to sporadic test failures.
2018-02-02smpschedaffinity04: Fix configurationSebastian Huber1-0/+2
2018-02-02smpcapture01: Fix configurationSebastian Huber1-0/+2
2018-01-25Remove make preinstallChris Johns1-0/+2
A speciality of the RTEMS build system was the make preinstall step. It copied header files from arbitrary locations into the build tree. The header files were included via the -Bsome/build/tree/path GCC command line option. This has at least seven problems: * The make preinstall step itself needs time and disk space. * Errors in header files show up in the build tree copy. This makes it hard for editors to open the right file to fix the error. * There is no clear relationship between source and build tree header files. This makes an audit of the build process difficult. * The visibility of all header files in the build tree makes it difficult to enforce API barriers. For example it is discouraged to use BSP-specifics in the cpukit. * An introduction of a new build system is difficult. * Include paths specified by the -B option are system headers. This may suppress warnings. * The parallel build had sporadic failures on some hosts. This patch removes the make preinstall step. All installed header files are moved to dedicated include directories in the source tree. Let @RTEMS_CPU@ be the target architecture, e.g. arm, powerpc, sparc, etc. Let @RTEMS_BSP_FAMILIY@ be a BSP family base directory, e.g. erc32, imx, qoriq, etc. The new cpukit include directories are: * cpukit/include * cpukit/score/cpu/@RTEMS_CPU@/include * cpukit/libnetworking The new BSP include directories are: * bsps/include * bsps/@RTEMS_CPU@/include * bsps/@RTEMS_CPU@/@RTEMS_BSP_FAMILIY@/include There are build tree include directories for generated files. The include directory order favours the most general header file, e.g. it is not possible to override general header files via the include path order. The "bootstrap -p" option was removed. The new "bootstrap -H" option should be used to regenerate the "headers.am" files. Update #3254.