#16 Resolve lack of SDK resources

Closed
opened 3 years ago by Michael Schloh von Bennewitz · 1 comments

Resolve lack of SDK resources

Problem environment

SDK and development environment allow for rich workflows with critical libraries, examples, programming, debugging, and documentation. For this reason testing of SDK resources is a valuable use of time.

Steps to reproduce

  1. Test PlatformIO
  2. Test MBed
  3. Test others

Expected result

Combinations of SDK, IDE, and RF software stack (library support) is aligned, so that one set of portable tools can lead the developer in a workflow concluding in running embedded code.

Actual result

Although Mbed and Zephyr platform targets (for compiling) exist, only limited library support (for Arduino) exists. This makes development difficult at best, and guarantees errors, bugs, and extra porting work.

Workaround

Microchip architecture

Atmel Studio seems to have a full featured environment, as well as a full LoRaWAN software stack (the missing libraries.) The environment has poor portability, and few open interfaces.

STMicro architecture

STM32CubeMX has most desirable features, with less open interfaces and rich library choice than PlatformIO. Unfortunately, the STM RF part is too new to be included, and the ST software stack for LoRa is too new to be acurately judged.

Severity level

This is medium priority That is easy to complete given enough time, or extremely difficult to complete given limited time.

# Resolve lack of SDK resources ## Problem environment SDK and development environment allow for rich workflows with critical libraries, examples, programming, debugging, and documentation. For this reason testing of SDK resources is a valuable use of time. ## Steps to reproduce 1. Test PlatformIO 2. Test MBed 3. Test others ## Expected result Combinations of SDK, IDE, and RF software stack (library support) is aligned, so that one set of portable tools can lead the developer in a workflow concluding in running embedded code. ## Actual result Although Mbed and Zephyr platform targets (for compiling) exist, only limited library support (for Arduino) exists. This makes development difficult at best, and guarantees errors, bugs, and extra porting work. ## Workaround ### Microchip architecture Atmel Studio seems to have a full featured environment, as well as a full LoRaWAN software stack (the missing libraries.) The environment has poor portability, and few open interfaces. ### STMicro architecture STM32CubeMX has most desirable features, with less open interfaces and rich library choice than PlatformIO. Unfortunately, the STM RF part is too new to be included, and the ST software stack for LoRa is too new to be acurately judged. ## Severity level This is **medium priority** That is easy to complete given enough time, or extremely difficult to complete given limited time.
Michael Schloh von Bennewitz added the
wontfix
label 3 years ago
Michael Schloh von Bennewitz added the
bug
label 3 years ago
Michael Schloh von Bennewitz added this to the Early testing against existing SDK milestone 3 years ago

After testing a number of tools, we conclude that no combination exists to allow us the flexibility and rich featureset needed. The hope is that given a month or two, tools begin to incorporate ST Micro software. Until then a workaround must be found.

After testing a number of tools, we conclude that no combination exists to allow us the flexibility and rich featureset needed. The hope is that given a month or two, tools begin to incorporate ST Micro software. Until then a workaround must be found.
Sign in to join this conversation.
No Assignees
1 Participants
Due Date

No due date set.

Dependencies

This issue currently doesn't have any dependencies.

Loading…
Cancel
Save
There is no content yet.