A8F Timing Benchmarks

ti.platforms.evmDM8168

Benchmark Cycles (1)
Interrupt latency 1746
Hwi_enable 15
Hwi_disable 22
Hwi dispatcher prolog 1812
Hwi dispatcher epilog 758
Hwi dispatcher 940
Hardware Interrupt to Blocked Task 1655
Hardware Interrupt to Software Interrupt 1910
Swi_enable 497
Swi_disable 307
Post Software Interrupt Again 93
Post Software Interrupt without Context Switch 309
Post Software Interrupt with Context Switch 291
Create a New Task without Context Switch 2179
Set a Task Priority without a Context Switch 688
Task_yield 860
Post Semaphore, No Waiting Task 95
Post Semaphore No Task Switch 348
Post Semaphore with Task Switch 444
Pend on Semaphore, No Context Switchi 218
Pend on Semaphore with Task Switch 362
Clock_getTicks 13

(1) The benchmark application was built using BIOS.LibType_Custom with the following compiler options: "--neon --endian=little -mv7A8 --abi=eabi -ms --embed_inline_assembly --opt_for_speed=2 --program_level_compile -o3 -g --optimize_with_debug".

Timings were obtained using the evmDM8168 development board.

The A8 core was running at 987MHz, with L1 & L2 caches enabled, and all code & data placed in OCMC0.

Unlike other target benchmarks that are collected using flat memory simulators or hardware with zero wait-state memory and no cache, these A8F numbers are collected on a real board with cache enabled and no-attempt to pre-fill the cache. These numbers are effected by varying amounts of cache filling depending on the dynamics of the cache and the previous function calls.