Crash on & after Read; inconsistent error messages & timing

[zacharyweiss]

Howdy! Thanks for your work on this; it's an exceptionally exciting project.

Just received and wired up a YRM1001 [the "1dbi 25x35mm" version from the link in the README] for the first time today, testing reads on the assortment of tags they shipped with the YRM. First couple read without issue, but since I've been primarily getting errors. The following were received after the scene had successfully switched to uhf_scene_read_tag but before switching to uhf_scene_read_tag_success; presumably upon detecting the tag (based on timing alone).

• "BusFault"
• "[something...], possibly stack overflow"
• "UsageFault"

Additionally, "HardFault" was recieved moments after switching to uhf_scene_read_tag_success populated with data, with no additional user input between the scene change and the fault.

Module Info as reported by the app [sic, the "`" is present in the readout]:

HW Version: M100 26dBm V1.0
SW Version: V2.3.5X
Manufacturer: MagicRf`

Device info:
FW: MNTM-DEV CACDC684 [shouldn't have any functional differences from MNTM-001, to my knowledge]
HW: RGB mod; data spliced to VIBRO test pad [color set to static white, so unlikely to be eating up anything in the background]
Log Level: None

I'd offer trace logs, but the software locks me out of 5V on GPIO while a USB is connected; fine to force? Already receiving 5V by proxy from the USB, w/o the 5V on GPIO option set?

Any thoughts on what might be causing this?

[Didgitalpunk]

(no clue, been getting the same, but it's still a work in progress app so I think that's to be expected)
what firmware version and app release are you using?

[zacharyweiss]

FW: MNTM-DEV CACDC684 [shouldn't have any functional differences from MNTM-001, to my knowledge]

App version is the one currently bundled in Momentum-Apps; last merge seems to have been a month ago. Shall I clone & build the most recent version here?

EDIT: Versioning a bit hard to track as the bundled MNTM version is just merging in the Unleashed one, and for these the commits are all associated to / authored by xMasterX rather than being a fork. It appears however that the version linked is two commits ahead of the current master branch one here, with a view alloc/free fix & an unintended-UART-deinit fix added.

[Didgitalpunk]

I suggest you take it up with the maker(s) of the CFW you're using then, especially if they made changes to the code

[zacharyweiss]

Right you are; flashed OFW commit 12112e7 (the current dev branch) & the current master branch here (commit a842673). From very quick testing: no crashes, only freezes, as described in #12. Will close this, and follow up with them. Thanks!