Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How to use dma_afu example for processing data #48

Open
andrea-ca opened this issue Nov 20, 2019 · 2 comments
Open

How to use dma_afu example for processing data #48

andrea-ca opened this issue Nov 20, 2019 · 2 comments
Assignees

Comments

@andrea-ca
Copy link

Hi,
I using the card intel PAC arria 10 gx FPGA(https://www.intel.com/content/dam/www/programmable/us/en/pdfs/literature/po/intel-pac-with-arria-10-gx-product-brief.pdf).
My goal is to create an FPGA accelerator that uses the DMA for the
transaction of data between host and local memory of the FPGA. To do
this, I use the example "dma_afu" present on the Github (OPAE)
website. I want to insert a processing module inside the example.
I guess to add a module inside the "msgdma_bbb" block in this manner:
the processing module is put between "the dma_read_master" and
"dma_write_master", and connected to them through the avalon_st interface.
This additional module must include an enabling signal which, if it is
off, will not make any changes to the data (normal operation) and if it
is on, the data will be processed.
My problem is that I can't find a signal that can be used for this purpose.
I was suggested to use the csr_slave signal used in the dispatcher but I
do not know to do this. Any suggestion?
Thanks.
Best regard

@jarrodblackburn
Copy link

My recommendation is to use the streaming DMA sample design instead. The streaming DMA design example contains a pair of memory-to-stream and stream-to-memory DMA engines that expose Avalon-ST interface that your own block can sit between. This way you perform a pair of DMA transfers to move data to and from the host and the data flows through your block performing some sort of transformation of the data.

Currently in the streaming DMA design there are pattern checkers and generators connected to each DMA engine. You could remove those and place your block between the DMAs if your block is already Platform Designer ready, or just export the streaming interface of each DMA engine and integrate your block in RTL further up in the design hierarchy.

@jarrodblackburn
Copy link

jarrodblackburn commented Nov 20, 2019

Here is a link to the streaming DMA user guide for the design that accompanies the 1.2 release of the A10 PAC board: https://www.intel.com/content/dam/altera-www/global/en_US/pdfs/literature/ug/ug-st-dma-afu.pdf

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants