summaryrefslogtreecommitdiffstats
path: root/cpukit/libtrace (follow)
Commit message (Collapse)AuthorAgeFilesLines
* record: Add rtems_record_dump()Sebastian Huber2020-03-185-0/+480
| | | | | | | | | Add rtems_record_dump_base64() and rtems_record_dump_base64_zlib(). Add CONFIGURE_RECORD_FATAL_DUMP_BASE64 and CONFIGURE_RECORD_FATAL_DUMP_BASE64_ZLIB configuration options. Update #3904.
* Use RTEMS_SYSINIT_ORDER_LAST_BUT_5Sebastian Huber2020-02-041-1/+1
| | | | | | | | Use RTEMS_SYSINIT_ORDER_LAST_BUT_5 instead of RTEMS_SYSINIT_ORDER_LAST to allow applications and support functions to place system initialization handlers behind the standard handlers. Update #3838.
* Revert "record: Add wrappers for malloc() functions"Sebastian Huber2019-10-021-145/+0
| | | | | | It was accidentally committed. This reverts commit a314544a278f1533ae001f742d7fe24fcee253e6.
* record: Add wrappers for malloc() functionsSebastian Huber2019-10-011-0/+145
| | | | | | | | | | Introduce new library librtemsrecordwrap.a which contains wrappers for operating system functions which produce entry/exit events. The wrappers can be selected during link time via the GNU ld --wrap option. Update #3665.
* record: Add system eventsSebastian Huber2019-09-021-3/+3
| | | | Update #3665.
* record: Add system eventsSebastian Huber2019-08-302-30/+570
| | | | | | Add system events for memory allocation/free. Update #3665.
* record: Add system eventsSebastian Huber2019-08-304-42/+129
| | | | | | | Add system events to identify the target system. Add system events to transfer blocks of memory and register sets. Update #3665.
* record: Simplify configurationSebastian Huber2019-08-301-4/+8
| | | | Update #3665.
* record: Introduce _Record_Drain()Sebastian Huber2019-08-301-36/+44
| | | | | | This allows its use in crash dump procedures. Update #3665.
* record: Add more system eventsSebastian Huber2019-08-292-64/+768
| | | | | | | | | | Reduce the system dependencies to allow tracing of very low level functions, for example the interrupt disable/enable. Introduce general purpose RTEMS_RECORD_CALLER and RTEMS_RECORD_LINE events. Update #3665.
* record: Introduce <rtems/recordserver.h>Sebastian Huber2019-08-281-0/+1
| | | | | | | This helps to get rid of the <rtems/rtems/tasks.h> dependency in <rtems/record.h>. Update #3665.
* record: Use BSS section instead of per-CPU dataSebastian Huber2019-08-282-8/+8
| | | | | | | | | | | The .rtemsrwset section is used for the per-CPU data. This section has loadable content. Place the ring buffers in the BSS section to avoid large executable image sizes. Not using the per-CPU data makes it possible to initialize the record support earlier. Update #3665.
* record: Add data size to clientSebastian Huber2019-08-271-0/+8
| | | | | | | This is necessary to get the thread names properly on 32-bit and 64-bit targets. Update #3665.
* record: Fix thread names on 64-bit targetsSebastian Huber2019-08-272-2/+8
| | | | | | Also fixes the thread names on signed char targets. Update #3665.
* record: Improve overflow handlingSebastian Huber2019-08-261-25/+58
| | | | | Signal the accumulated item overflow count with the time of the first new item.
* record: Fix off by one errorSebastian Huber2019-08-261-1/+1
|
* record: Pass bintime to client handlersSebastian Huber2019-08-171-8/+1
| | | | This is a minor optimization.
* record: Simplify client visit()Sebastian Huber2019-08-171-14/+24
|
* record: Remove superfluous castSebastian Huber2019-08-171-1/+1
|
* record: Increase client robustnessSebastian Huber2019-08-171-10/+33
| | | | Do nothing after errors.
* record: Improve overflow handlingSebastian Huber2019-08-171-9/+198
| | | | | | | | In case of a ring buffer overflow, the rtems_record_drain() will push the complete ring buffer content to the client. While the items are processed by the client, new items may overwrite some items being processed. The overwritten items can be detected in the following iteration once the next tail/head information is pushed to the client.
* record: Detect also large overflowsSebastian Huber2019-08-161-18/+33
|
* record: Simplify rtems_record_client_contextSebastian Huber2019-08-161-17/+27
|
* record: Change thread name encodingSebastian Huber2019-08-062-2/+2
| | | | This scheme is easier to decode.
* record: Add support for thread namesSebastian Huber2019-07-303-6/+102
|
* rtems: Add rtems_scheduler_get_processor_maximum()Sebastian Huber2019-04-091-1/+1
| | | | | | | | | | | 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.
* record: Add more system eventsSebastian Huber2019-03-121-57/+57
| | | | Update #3665.
* record: Add system call entry/exit eventsSebastian Huber2019-03-121-82/+82
| | | | | | This corresponds to the Linux syscall_entry_* and syscall_exit_* events. Update #3665.
* record: Rename internal per-CPU eventsSebastian Huber2019-03-124-11/+11
| | | | Update #3665.
* record: Add enum value for each eventSebastian Huber2019-01-302-10/+893
| | | | Update #3665.
* Add low level event recording supportSebastian Huber2019-01-296-0/+1320
Add low level event recording infrastructure for system and user defined events. The infrastructure is able to record high frequency events such as * SMP lock acquire/release, * interrupt entry/exit, * thread switches, * UMA zone allocate/free, and * Ethernet packet input/output, etc. It allows post-mortem analysis in fatal error handlers, e.g. the last events are in the record buffer, the newest event overwrites the oldest event. It is possible to detect record buffer overflows for consumers that expect a continuous stream of events, e.g. to display the system state in real-time. The implementation supports high-end SMP machines (more than 1GHz processor frequency, more than four processors). Add a new API instead. The implementation uses per-processor data structures and no atomic read-modify-write operations. It is uses per-processor ring buffers to record the events. The CPU counter is used to get the time of events. It is combined with periodic uptime events to synchronize it with CLOCK_REALTIME. The existing capture engine tries to solve this problem also, but its performance is not good enough for high-end production systems. The main issues are the variable-size buffers and the use of SMP locks for synchronization. To fix this, the API would change significantly. Update #3665.
generated by cgit v1.2.3 (git 2.25.1) at 2024-04-27 20:06:45 +0000