Hello World doesnt print message to JTAG uart console

[JayHarikumaran]

Hello,

I am trying run the code from Flash. So I created a PoF file using the .sof file and the hex file for the flash as shown in https://www.youtube.com/watch?v=0k4AZmdW9Sk to run code from flash.

When I launch run as Nios II I dont see any Hello world message on the Nios II console. I am using the hello world free standing template.

I have attached the qsys design file as well as the qsf file for the top project. I am using a DE-10 lite board to try the design.

Regards

Jay

[JayHarikumaran]

I was able to step through the code and the code is executing each statement but the printf is not sending the message onto the console.

[JayHarikumaran]

I saw on another thread about disabling c_plus_plus in BSP.
https://community.intel.com/t5/Nios-II-Embedded-Design-Suite/Processor-stuck-at-reset/m-p/151894

Tried that also unsuccessfully. The error I am getting now is:

"Reading symbols from C:\FPGA_simulations\DE-10\Boulder\Nios1\software\HelloFree_Flash\HelloFree_Flash.elf...
_start () at HAL/src/crt0.S:304
304 movhi sp, %hi(__alt_stack_pointer)
Warning:
Cannot insert breakpoint 1.
Cannot access memory at address 0x80270"

The above memory location is in Flash and I am trying to run from flash.

Here is a screenshot of the Qsys system. The incoming clock on DE-10 is 50Mhz and I am using the PLL to generate 100MHz on which all the components are running.

I have added the clock constraints alone for timing without adding an input/output delay as at this point, I am not doing anything accessing external logic.

[JayHarikumaran]

I have attached the Qsys design file. The dual_boot ip allows maximum clock rate of 80MHz. I have updated that as well in case that would be the issue and it still throws up the same error.

[JayHarikumaran]

I changed the reset vector for Nios II to Ram and increased ram size as the application plus stack and heap wouldnt fit in the RAM from 32KB to 66000 bytes and kept everything else the same. Recreated BSP to run code from RAM and the processor does work. I think the UFM should be large enough to hold the same program and data plus the copier functions. Could someone share their thoughts on this issue?

How

[EricMunYew_C_Intel]

Have you tried alt_printf, it can reduce the UFM memory requirement.