95a97fd287
Ports the Thread-Metric suite of benchmarks from ThreadX to Zephyr. This makes it easier for others to run these benchmarks with the best set of configuration options for their board so that they can better compare Zephyr performance to another RTOS. Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
84 lines
2.7 KiB
Plaintext
84 lines
2.7 KiB
Plaintext
# Copyright (c) 2024 Intel Corporation
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
|
|
mainmenu "Eclipse ThreadX Thread-Metric RTOS Test Suite"
|
|
|
|
source "Kconfig.zephyr"
|
|
|
|
choice TM_TEST
|
|
prompt "Select a Thread-Metric test to execute"
|
|
default TM_PREEMPTIVE
|
|
help
|
|
The Thread-Metric benchmark suite consists of eight RTOS tests.
|
|
These tests measure the total number of RTOS events that can be
|
|
processed during a 30 second time interval.
|
|
|
|
config TM_BASIC
|
|
bool "Baseline basic benchmark"
|
|
help
|
|
The baseline basic benchmark consists of a single thread that counts
|
|
the number of times it performs a set of calculations. This number
|
|
is reported every 30 seconds.
|
|
|
|
config TM_COOPERATIVE
|
|
bool "Cooperative context switching"
|
|
help
|
|
The cooperative context switching benchmark spawns five (5) threads
|
|
of equal priority that yield to each other and increment counters
|
|
on each context switch. The sum total of the counters is reported
|
|
every 30 seconds.
|
|
|
|
config TM_INTERRUPT
|
|
bool "Interrupt processing"
|
|
select TEST
|
|
select IRQ_OFFLOAD
|
|
select IRQ_OFFLOAD_NESTED
|
|
help
|
|
The interrupt processing benchmark has a single thread that causes
|
|
an interrupt which results in its ISR incrementing a counter and then
|
|
posting a semaphore. The thread then increments its own counter and
|
|
takes that semaphore. The sum total of the counters is reported
|
|
every 30 seconds.
|
|
|
|
config TM_INTERRUPT_PREEMPTION
|
|
bool "Interrupt processing preemption"
|
|
select TEST
|
|
select IRQ_OFFLOAD
|
|
select IRQ_OFFLOAD_NESTED
|
|
help
|
|
The interrupt preemption benchmark counts the number of times that
|
|
an ISR from a software generated interrupt results in the preemption
|
|
of a thread. The total number of context switches is reported every
|
|
30 seconds.
|
|
|
|
config TM_MEMORY_ALLOCATION
|
|
bool "Memory allocation"
|
|
help
|
|
The memory allocation benchmark counts the number of times a thread
|
|
is able to allocate and then release a 128-byte block. This number
|
|
is reported every 30 seconds.
|
|
|
|
config TM_MESSAGE
|
|
bool "Message processing"
|
|
help
|
|
The message processing benchmark counts the number of times that a
|
|
thread can send and receive a 16-byte message from a message queue.
|
|
This number is reported every 30 seconds.
|
|
|
|
config TM_PREEMPTIVE
|
|
bool "Preemptive context switching"
|
|
help
|
|
The preemptive context switching benchmark creates five (5) threads
|
|
of different priorities that suspend and resume each other in a
|
|
cyclical pattern. The total number of context switches is reported
|
|
every 30 seconds.
|
|
|
|
config TM_SYNCHRONIZATION
|
|
bool "Synchronization"
|
|
help
|
|
The synchronization benchmark counts the number of times that a
|
|
thread can give and take a semaphore without blocking. This number
|
|
is reported every 30 seconds.
|
|
|
|
endchoice
|