• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 98 subscribers
  • Views 356 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

urgent question about hyperlink example

zezo alex
Prodigy 170 points

hello ;

i am using CCS 5.0.3 , C6670 lite evolution module , windows 7

i am trying to use hyperlink example in the following directory:

C:\TI\pdk_C6670_1_0_0_17\packages\ti\drv\exampleProjects\hyplnk_exampleProject, i successfully debug the example but i have 2 questions

1-how to exactly use the memory viewer and if there is an illustrative link?
2- what is exactly the role of iterations and the tokens??
regarding to the second question : i changed the value of hyplnk_EXAMPLE_NITERS - in line 161-(hyplnkExample.c)  to be 2 in order to have two iterations only and i changed the value of :hyplnk_EXAMPLE_NTOKENS in line-156-(hyplnkExample.c) and i noticed that its value appeared in certain location in the segments .


 

  • 0
    TI__Expert 4920 points

    Hi Zezo

    The memory viewer is explained at http://processors.wiki.ti.com/index.php/GSG:Debugging_projects_v5#Memory_viewer

    If you look at the function hyplnkExampleIOCycle that is called within the 'for' loop, you'll notice that this function first does a CPU token exchange, i.e. it passes one "token" by writing to the "write" address then attempting to read the same value back via the "read" address. It then does a CPU block transfer, i.e. it writes a block of data to the remote side then verifies it. If you look at the structure hyplnkExampleOneToken_t that represents a "token", you'll notice that a token is a 32-bit value followed by a padding of 127 bytes to align to 128 bytes. The buffer that is exchanged for the memory block transfer is an array of 4096 32-bit values.

    The hyplnk_EXAMPLE_NTOKENS just represents the number of tokens and buffers that will be part of the above hyperlink test. The profiling is also done based on this number, that is 'cycles-per-roundtrip = total_time/hyplnk_EXAMPLE_NTOKENS' which provides an average value of time taken for the token and block transfer.

    The iterations are just the number of times you run the test. By default, the test is set to run forever. You can set a lower number of iterations if you choose. Each iteration will run the read-write test hyplink_EXAMPLE_NTOKENS and provide an average time.

    Note that the address translation setup for hyperlink is done before the 'for' loop so that will not change from one iteration to the next.

  • 0 in reply to Uday
    Prodigy 170 points

    thanks ran for your reply ,but i want to do the following

    i have a code (i made it in c-language) , there is a certain i/p to my code(stream of ones and zeros) and the expected o/p of my code is stream of ones and zeros too , i want the processor to process code then the o/p is sent to the hyperlink interface then sending the o/p of my code to the remote device (which is a stream of bits) . After the remote device receive the stream of ones and zeros,there is another code in the remote device taking the stream of ones and zeros and the processor process code then displaying the o/p.

    So,i want to send a stream ones&zeros  from one side to the other side  and as i ran the example,the example send a one token value,so can you please help me in doing what i want to do as soon as possible??

    Regards