Yes, the issue was resolved once I did some more work on it.
Note that my issue was bigger than JUST the definition of the platform: I also needed a way to get the filesystem onto the flash without an FTDI debugger present.
This is the latest version of the SDK, 6.1.0 .
So my implementation is a bit roundabout:
1) Build any project for my platform using `RESOURCES_LOCATION ?= RESOURCES_IN_WICEDFS` in the platform makefile
2) This causes the build system to generate filesystem.bin in the build/ directory; this BIN contains the wifi FW for the platform and the CLM blob.
3) Use `./tools/text_to_c/bin_to_resource_c.pl` to convert filesystem.b into a C file of the binary data.
4) Author a custom app that includes the filesystem.c file. (Note: I couldn't get the definition as an extern to work, so I ended up having to #include the C file)
5) The custom app constructs the headers directly by filling out a `app_header_t headers[DCT_MAX_APP_COUNT]` array, and then flashing it to sector 0 of the flash. This includes correct size provisioning for the filesystem-as-a-variable.
6) In the section corresponding to DCT_FILESYSTEM_IMAGE_INDEX, the filesystem variable is written out to the flash section.
7) Contents of the DCT_FILESYSTEM_IMAGE_INDEX are read back and verified against the filesystem-as-a-variable data.
AFTER that:
😎 I had the filesystem in place in the flash, but the system was still not working, I hit a timeout in `wwd_bus_sdio_download_firmware`, line 827 of wwd_bus_protocol.c
9) Upon review, I found that platform definitions now require `#define USES_RESOURCE_FILESYSTEM` in the `platform_config.h`
10) After adding that macro and doing the above preparation for the sflash, the platform can load correctly and start the radio for a project with the same RESOURCES_IN_WICEDFS macro.
This wasn't easy. But was necessary, since it meant the difference between a 1MB chip and a 1.5MB .
To address steps 1-7, at some point in the future I plan to write a program that runs a softAP and webserver or simple TCP that will let me remotely edit the sflash header, write any contents by streaming, and then retrieve a BIN of the flash.
Maybe this could be a community effort? Or Cypress could get an intern to do it? That would be a great intern project 
