Closed filzek closed 6 months ago
There are a few time/cpu expensive operations within littlefs
. It seems like your current issue is that some littlefs
operations take up too much time, and consequently trigger your watch dog timers?
If so, I suppose there are a few options:
CONFIG_ESP_TASK_WDT_TIMEOUT_S
.taskYIELD()
before filesystem operations.lfs_fs_gc
, that can front-load some of the expensive operations. I could add an interface to expose this, and the user could call this during times the CPU would otherwise be idling.did not solve the problem, the system got corrupted everywhere in 1.10.1, we will try the 1.10.2 version
Can you explain your usage context a bit more? Also have you tried debugging to see what is causing your system to hang?
Closing due to lack of activity.
Hi,
we are using the latest version of 1.10.1 with idf 4.4.6 and when we replace the FatFS with LittleFS the system dispare a lot of wdts everywhere, so, something is intense consuming cpu usage during flash operations, we have to add a lot of task wait to clear the wdts.
Is that normal or a buggy?
In the FatFS everything works perfectly without any wdt.
The LittleFS doesnt dispare any WDT reset only intense wdt usage.