2021-05-26 08:46 AM
Hi,
I've some questions about validation on target, we meet some issues.
Two ways are offer to do that, with or without automatic generate, compile, etc.
1 - We tried the second method, and code generation stop due to not enough internal RAM. We are not able to modify STM32H745XI.ld to change AXI-SRAM lenght from 256KB to 512KB due to automatic code generation and we don't know local directory used for this method.
Can we make a workarround if we decide to use external SRAM for neural network ?
2 - We tried first method and build, program targets following AN5361 for configuration.
We build project as debug application, however CubeMX can't find eval board for validation on target.
3 - Last question about performance validation, we only select "Performance" on X-Cube-AI Device Application menu ?
Thanks you for your support,
BR
Benjamin
2021-05-31 01:03 AM
Hi community,
Someone can help us ?
We tried with validation and performance application, same results.
No communication between board and CubeMX or TeraTerm.
BR,
Benjamin
2021-06-03 12:00 PM
Hi,
After few tests i' able to :
So maybe I brick M4 access port or STlink ?
I will appreciate a support from STemployee if possible... or someone else who meet this situation.
BR
Benjami