THREADX RTOS Real-Time Operating System
Most deployed RTOS – over billion deployments!
The popularity of THREADX is a testament to its reliability, quality, size, performance, advanced features, ease-of-use, and overall time-to-market advantages.
“We have followed the growth trajectory of THREADX in the wireless and IoT markets since the company’s founding, and are increasingly impressed by the widespread industry adoption of THREADX.” – Chris Rommel, Executive Vice President, VDC Research
picokernel™ architecture and automatic scaling. Automatic scaling means that only the services (and supporting infrastructure) used by the application are included in the final image at link time.
Here are some typical THREADX size characteristics:
|THREADX Service||Typical Size in Bytes|
|Core Services (Require)||2,000|
|Event Flag Services||900|
|Block Memory Services||550|
|Byte Memory Services||900|
Fast, deterministic execution
commercial RTOSes. In addition to being fast, THREADX is also highly deterministic – it achieves the same fast performance whether there are 200 threads ready, or just one.
Here are some typical performance characteristics of THREADX:
- Fast Boot: THREADX boots in less than 120 cycles.
- Optional Removal of basic error checking: Basic THREADX error checking can be skipped at compile time. This can be useful when the application code is verified and no longer requires error checking on each parameter. Note that this can be done on a compilation unit, rather than system-wide.
- Picokernel™ Design: Services are not layered on each other, thus eliminating unnecessary function call overhead.
- Optimized Interrupt Processing: Only scratch registers are saved/restored upon ISR entry/exit, unless preemption is necessary.
- Optimized API Processing:
|THREADX Service||Service Time in Microseconds|
|Interrupt Response||0.0 – 0.6|
Performance figures based on typical processor running at 200MHz.
For a comprehensive data sheet with size and performance estimates for your particular architecture, contact us at .
Pre-certified by TUV and UL to many safety standards
SW Safety Class C, ISO 26262 ASIL D and EN 50128. The certification confirms that THREADX and THREADX SMP can be used in the development of safety-related software for the highest safety integrity levels of IEC-61508, IEC-62304, ISO 26262 and EN 50128 for the “Functional Safety of electrical, electronic, and programmable electronic safety-related systems.” SGS-TUV Saar, formed through a joint venture of Germany’s SGS-Group and TUV Saarland, has become the leading accredited, independent company for testing, auditing, verifying and certifying embedded software for safety-related systems worldwide. The industrial safety standard IEC 61508, and all standards that are derived from it, including IEC-62304, ISO 26262 and EN 50128, are used to ensure the functional safety of electrical, electronic, and programmable electronic safety-related medical devices, process control systems, industrial machinery, automobiles and railway control systems.
THREADX and THREADX SMP have been recognized by UL for compliance with UL 60730-1 Annex H, CSA E60730-1 Annex H, IEC 60730-1 Annex H, UL 60335-1 Annex R, IEC 60335-1 Annex R, and UL 1998 safety standards for software in programmable components. UL is a global, independent, safety-science company with more than a century of expertise innovating safety solutions, ranging from the public adoption of electricity to breakthroughs in sustainability, renewable energy, and nanotechnology.
Artifacts (Certificate, Safety Manual, Test Report, etc.) associated with the TUV and UL certifications are available for sale. Please contact us for more details at .
In cases where the application needs additional certification, a certification service is available through Express Logic to provide turn-key certification to various standards, using the actual hardware platform and even covering the application code. Please contact us for more details on our certification service.
words and not the “alphabet soup” or highly abbreviated names that are so common in other RTOS products. All THREADX APIs have a leading “tx_” and follow a noun-verb naming convention. Furthermore, there is a functional consistency throughout the API. For example, all APIs that suspend have an optional timeout that functions in an identical manner for APIs.
Building a THREADX application is easy. The application needs to include tx_api.h, call tx_kernel_enter from main, define the tx_application_define function and create one thread, define the thread entry point function, and link against the THREADX library (typically tx.a). That simple set of requirements is best illustrated with this simple 26-line THREADX code:
THREADX also boasts the highest caliber of documentation available. View our THREADX User Guide and see for yourself!
and has been the subject of extensive academic research, including:
“Scheduling Fixed-Priority Tasks with Preemption Threshold,” by Yun Wang, Concordia University, and Manas Saksena, University of Pittsburgh; www.cs.utah.edu/~regehr/reading/open_papers/preempt_thresh.pdf
“Using Preemption-Threshold Scheduling to Cut Overhead While Meeting Deadlines,” by Dr. Alexander G. Dean, Center for Efficient, Secure and Reliable Computing, Dept. of Electrical and Computer Engineering, North Carolina State University.
Consider the capabilities of THREADX technology:
- Complete and Comprehensive Multitasking Facilities
- Threads, application timers, message queues, counting semaphores, mutexes, event flags, block and byte memory pools
- Priority-Based Preemptive Scheduling
- Priority Flexibility – Up to 1024 Priority Levels
- Cooperative Scheduling
- Preemption-Threshold™ – Unique to THREADX, helps reduce context switches and help guarantee schedulability (per academic research)
- Memory Protection via THREADX MODULES
- Fully Deterministic
- Event Trace – Capture last “n” system/application events
- Event Chaining™ – Register an application-specific “notify” callback function for each THREADX communication or synchronization object
- THREADX MODULES with Optional Memory Protection
- Run-Time Performance Metrics
- Number of thread resumptions
- Number of thread suspensions
- Number of solicited thread preemptions
- Number of asynchronous thread interrupt preemptions
- Number of thread priority inversions
- Number of thread relinquishes
- Execution Profile Kit (EPK)
- Separate Interrupt Stack
- Run-Time Stack Analysis
- Optimized Timer Interrupt Processing
Multicore support (AMP & SMP)
and the application (or Linux) execute on each core and communicate with each other via shared memory or an inter-processor communication mechanism such as OpenAMP (THREADX supports OpenAMP). This is the most typical multicore configuration using THREADX and can be the most efficient if the application is able to effectively load the processors.
For environments where loading the processors is highly dynamic, THREADX Symetric Multiprocessing (SMP) is available for the following processor families:
- ARM Cortex-Ax
- ARM Cortex-Rx
- ARM Cortex-A5x 64-bit
- MIPS 34K, 1004K, and interAptiv
- Synopsys ARC HS
THREADX SMP performs dynamic load balancing across “n” processors and allows all THREADX resources (queues, semaphores, event flags, memory pools, etc.) to be accessed by any thread on any core. THREADX SMP enables the complete THREADX API on all cores and introduces the following new API’s applicable to SMP operation:
- UINT tx_thread_smp_core_exclude(TX_THREAD *thread_ptr, ULONG exclusion_map);
- UINT tx_thread_smp_core_exclude_get(TX_THREAD *thread_ptr, ULONG *exclusion_map_ptr);
- UINT tx_thread_smp_core_get(void);
- UINT tx_timer_smp_core_exclude(TX_TIMER *timer_ptr, ULONG exclusion_map);
- UINT tx_timer_smp_core_exclude_get(TX_TIMER *timer_ptr, ULONG *exclusion_map_ptr);
Please contact us for more information on THREADX Multicore support.
Memory protection via THREADX MODULES
that can be dynamically loaded and run (or executed in place) on the target.
Modules enable field upgrade, bug fixing, and program partitioning to allow large applications to occupy only the memory needed by active threads.
Modules also have a completely separate address space from THREADX itself. This enables THREADX to place memory protection (via MPU or MMU) around the Module such that accidental access outside the module will not be able to corrupt any other software component.
For more information on THREADX MODULES, please visit the THREADX MODULES page.
time-to-market RTOS for the last seven consecutive years, per the Embedded Market Forecasters (EMF) surveys. The surveys consistently show that 70% of designs using THREADX get to market on time – surpassing all other RTOSes.
The following are some reasons for our consistent time-to-market advantage:
- Quality Documentation – Please review our THREADX User Guide and see for yourself!
- Complete Source Code Availability
- Easy-to-Use API
- Comprehensive and Advanced Feature Set
- Broad 3rd Party Tools Integration – Especially IAR’s Embedded Workbench™
There are five basic types of licenses and the ability to create a custom license for unique situations. Please see the licensing page for more information.
Full, highest-quality source code
Throughout the years, THREADX source code has set the bar in quality and ease of understanding. In addition, the convention of having one function per file provides for easy source navigation. The following is an example of the THREADX tx_thread_identify.c function:
THREADX conforms to strict coding conventions, including the requirement that every line of C code has a meaningful comment. In addition, the THREADX source has been certified to the highest standards. Please contact us today for a full source evaluation.
guidelines for critical systems using the C programming language. The original MISRA C guidelines were primarily targeted toward automotive applications; however, MISRA C is now widely recognized as being applicable to any safety-critical application. THREADX is compliant with all “required” and “mandatory” rules of MISRA-C:2004 and MISRA C:2012. THREADX is also compliant with all but three “advisory” rules. Please contact us for more information.
Supports most popular architectures
including the following:
- Analog Devices: SHARC, Blackfin
- ARM: ARM7, ARM9, ARM11, Cortex-M0/M3/M4/M7/A5/A7/A8/A9/A52/A53/A57/R4/R5
- Cadence: Xtensa, Diamond
- CEVA: TeakLlte-lll
- EnSilica: eSi-RISC
- Imagination: MIPS32 4K, 24K, 34K, 1004K, MIPS64 5K, microAptiv, interAptiv, proAptiv, M-Class
- Intel: x36/Pentium, XScale, NIOS II, Cyclone, Arria 10
- Microchip: AVR32, ARM7, ARM9, Cortex-M3/M4/M7, SAM3/4/7/9/A/C/D/E/G/L/SV, PIC24/PIC32
- Microsemi: RISC-V
- NXP: LPC, ARM7, ARM9, PowerPC, 68K, i.MX, ColdFire, Kinetis Cortex-M3/M4
- Renesas: SH, HS, V850, RX, RZ, Synergy
- Silicon Labs: EFM32
- Synopsys: ARC 600, 700, ARC EM, ARC HS
- ST: STM32, ARM7, ARM9, Cortex-M3/M4/M7
- Tl: C5xxx, C6xxx, Stellaris, Sitara, Tiva-C
- Xilinx: MicroBlaze, PowerPC 405, ZYNQ
Supports most popular tools
has the most comprehensive THREADX kernel awareness available. Additional tool integration includes GNU (GCC), ARM DS-5/uVision®, Green Hills MULTI®, Wind River Workbench™, Imagination Codescape, Renesas e2studio, Metaware SeeCode™, NXP CodeWarrior, Lauterbach TRACE32®, TI Code-Composer Studio, CrossCore and all analog devices.