Support.Tulipp.eu

Hi Giacomo, what part of the UAV use-case would you like to have. 1. We have developed a simple avoidance algorithm, which takes a disparity map as input and computes an evasion maneuvre for the UAV (https://www.youtube.com/watch?v=pMDMTUCVwKc). This was developed in C/C++ and runs on the CPU. 2. Th...

Hi all,

we have also posted the question into the Xilinx Forum:

https://forums.xilinx.com/t5/SDSoC-Envi ... alse#M1199

Regards,
Boitumelo

Hi,

you mean to set XAxiVdma_DmaSetBufferAddr? I only set two. Need to check if it helps. Why would I need to also configure write channel, if I only need the read channel?

\Thanks.

Hi all, I am preparing for a demo in which I want to visualize the results at the frame rate at which they are produced. I have adjusted the HDMI Test project as Emilie suggested in http://support.tulipp.eu/viewtopic.php?f=8&t=87. Unfortunately due to the restart of the VDMA Module the whole scr...

Hi everyone,

I get the following error message when compiling my project: section `.bss' will not fit in region `ps7_ddr_0_S_AXI_BASEADDR'

Does this mean the platform doesn't have enough DDR memory available? I am using the HDMI Test example from the Starter Guide with DDR size of 1G.

\Boitumelo

Thanks :D There is an example of how to stream efficiently between functions in the: https://github.com/tulipp-eu/tulipp-tool-chain/tree/master/STHEM/SDSoC-Image-Processing-Library ... stream between 2 functions in 1 accelerator, since it is more efficient than building 2 accelerators and stream bet...

Hi everyone,

how do I tell SDSoC to directly connect two accelerated functions with a temporary buffer as data exchange instead of writing the intermediate data to DDR?

\Boitumelo

Hi, Does anybody have an idea on how to stream image data line by line into an accelerated function, instead of pixel by pixel? Here is what I currently have: #pragma SDS data mem_attribute(input_img:PHYSICAL_CONTIGUOUS|NON_CACHEABLE) #pragma SDS data mem_attribute(output_img:PHYSICAL_CONTIGUOUS|NON...

Hi all, is there a memory limit for functions that are to be moved to HW? When I try to compile our stereo algorithm for HW I get the following warning: WARNING: [SDSoC 0-0] Hardware function 'sgm_rgba' LUT resource estimate (108063) exceeds platform 'HDMIpfm' resource capacity (78600) I canceled th...

I see. Makes sense.

Thanks a lot.

\Boitumelo