mirror of
https://github.com/jomjol/AI-on-the-edge-device.git
synced 2025-12-08 12:36:52 +03:00
Show PSRAM usage (#2206)
* centralize PSRAM usage (application code only) * update logging * update logging * fix use after free * initialize buffer * free rgb_image before ussing it for new allocation * use wrapper function * switch log level to debug * . * undo adding free() calls * . * add names to all CImage instances * . * . * . * revert changes of stbi_image_free() with free_psram_heap() on the places where is is not in PSRAM * . * typos * typo * Added MQTT Outbox explanation/warning * added CONFIG_SPIRAM_USE_MEMMAP explanation --------- Co-authored-by: CaCO3 <caco@ruinelli.ch>
This commit is contained in:
@@ -8,6 +8,10 @@
|
||||
#include "esp_log.h"
|
||||
#include "esp_heap_caps.h"
|
||||
|
||||
/* Enable this to use the PSRAM for MQTT Publishing.
|
||||
* This saves 10 kBytes of RAM, see https://github.com/jomjol/AI-on-the-edge-device/pull/2113
|
||||
* However we can run into PSRAM fragmentation issues, leading to insufficient large blocks to load the model.
|
||||
* See https://github.com/jomjol/AI-on-the-edge-device/issues/2200 */
|
||||
#define USE_PSRAM
|
||||
|
||||
#ifdef CONFIG_MQTT_CUSTOM_OUTBOX
|
||||
|
||||
Reference in New Issue
Block a user