This is an automated email from the ASF dual-hosted git repository.
jerzy pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/mynewt-core.git
The following commit(s) were added to refs/heads/master by this push:
new 174494407 stm32/hal_watchdog: Fix timeout calculations
174494407 is described below
commit 174494407379406fe6adcf54566f300f9e8b1da4
Author: Jerzy Kasenberg <[email protected]>
AuthorDate: Sat May 18 16:38:28 2024 +0200
stm32/hal_watchdog: Fix timeout calculations
All STM32 devices use LSI for watchdog functionality.
LSI frequency varies between MCUs and it's approximation
is define by LSI_VALUE in stm32xxx_hal_conf.h.
Code was assuming 32768 which is ok for devices that have
LSI rate at 30kHz.
Some device has 37kHz others 40kHz it those cases calculated
time could be too much off resulting in reset by watchdog.
Now code uses LSI_VALUE adding 10% to avoid cases when LSI
runs faster then nominal.
---
hw/mcu/stm/stm32_common/src/hal_watchdog.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/hw/mcu/stm/stm32_common/src/hal_watchdog.c
b/hw/mcu/stm/stm32_common/src/hal_watchdog.c
index 421993fb4..649b6b6f1 100644
--- a/hw/mcu/stm/stm32_common/src/hal_watchdog.c
+++ b/hw/mcu/stm/stm32_common/src/hal_watchdog.c
@@ -29,7 +29,8 @@ hal_watchdog_init(uint32_t expire_msecs)
uint32_t reload;
/* Max prescaler is 256 */
- reload = 32768 / 256;
+ /* LSI is not very precise, assume 10% inaccuracy when calculating reload
value */
+ reload = LSI_VALUE * 11 / 2560;
reload = (reload * expire_msecs) / 1000;
/* Check to make sure we are not trying a reload value that is too large */
