- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi all, I have a question that has been bothering me for almost a week and hopefully someone could help me out.
Here is my plan, there are 2 RAMs in FPGA, one receives data from HPS, call it RAM0, then the output of RAM0 is added 2 and then sent to RAM1, RAM1 receives it and sends it back to HPS. The logic is very simple here, however the timing of RAM reads and writes is really a headache to me. Now I'm using tcl file to test through System Console. If I write the data of 88, 77, 66, 55, 44, 33, 22, 11 to address[0] -- address[7] in the RAM0, then add 2 to each elements, and then store the added data in the RAM1, the output readdata from RAM1 is 13, 8A, 79, 68, 57, 46, 35, 24 from adress[0] to address[7]. You can see that every element is added by two but the value of address[7] in RAM0 is written into address[0] in RAM1, the value of address[0] in RAM0 is written into adderss[1] in RAM1, etc. I've tried several ways to adjust the timing but it still doesn't work. The code and the System Console results have been attached.Link Copied
2 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi fannyfang,
The Cortex-A9 supports out-of-order execution. If you are executing a series of instructions that do not depend on the data from previous instructions, they may execute in an order other than what you have written in your code. You can read more about it in the TRM. Regards, Sue- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Yes it's true, now I have figured it out and it turns out the the writing to second RAM needs one more cycle to wait. Thanks for help!
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page