3bbe42f789
Test `tracing.transport.uart.async.test` was working only due sheer serendipity: one of the tracing strings checked, sys_trace_k_mutex_lock_exit, was never being registered on the tracing buffer. However, the check never got to check this string, so the test was passing. Why wasn't it being registered? Tracing buffer size was a bit small for the flurry of messages - increasing it solves the problem, and is what this patch does. Why did the check never got to the missing string? Tracing thread has a lower priority, so when the k_sleep of test thread expired, it was preempted. Indeed, if the sleep was changed by any number of ticks, the test would fail. Even changing order - and thus number - of instructions executed could make this test fail. While at it, fixed some typos and an imprecise commentary on code. Fixes: #84169 Signed-off-by: Ederson de Souza <ederson.desouza@intel.com>
7 lines
144 B
Plaintext
7 lines
144 B
Plaintext
CONFIG_ZTEST=y
|
|
CONFIG_TRACING=y
|
|
CONFIG_TRACING_TEST=y
|
|
CONFIG_TRACING_BACKEND_UART=y
|
|
CONFIG_IDLE_STACK_SIZE=4096
|
|
CONFIG_TRACING_BUFFER_SIZE=4096
|