Here are some fixes for rtla timerlat auto-analysis.
The first one zeroes the variable that accounts for the amount of
thread interference when a new cycle begins.
The second fixes an imprecision on the IRQ delay account when
the value is near 0.
The last one fixes the case when an IRQ interference for the
next activation is considered for the previous one.
Daniel Bristot de Oliveira (3):
rtla/timerlat_aa: Zero thread sum after every sample analysis
rtla/timerlat_aa: Fix negative irq delay
rtla/timerlat_aa: Fix previous IRQ delay for IRQs that happens after
thread sample
tools/tracing/rtla/src/timerlat_aa.c | 32 ++++++++++++++++++++++------
1 file changed, 26 insertions(+), 6 deletions(-)
--
2.38.1
|