-
Notifications
You must be signed in to change notification settings - Fork 267
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
reset for axi_stream_master #422
reset for axi_stream_master #422
Conversation
added reset output to use when the axi_stream_master is generating the reset. changed wait until idle construction to use a notify from the bus process so that we are sure all transactions (including the reset) have been handled. added checking of reset input on places where it is required so that a reset is handled properly in the master. added axi_stream_reset to vci. added reset input to axi_stream_slave.vhd. added basic test for reset functionality.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I like the addition of the reset to the master, but i would keep the reset generation out of it. Instead add a new component. which generates resets. (e.g. with synchronous asynchronous assert). This way it keeps a nice separation of functionality. In most of the designs I know the reset comes from some other source and not the master itself.
|
||
notify_bus_process_done <= '1'; | ||
wait until notify_bus_process_done = '1'; | ||
notify_bus_process_done <= '0'; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Also, I'am not sure, why this is needed. waiting for tvalid = '0' and aclk should be sufficient.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this was necessary to be able to issue a reset 'request' and follow it up with a wait_until_idle to wait for it to complete.
When just adding a areset_out_n = '1' there was a timing hole and the wait_until_idle would complete to early.
@go2sh Has a point. It would be more modular if there was a separate reset generator VC and the axi stream just had reset as an input? What do you say @olafvandenberg? |
Ok, I agree, sound likes a plan, I'll need some time for it though. |
replaced reset out / reset in, with single reset input. adapted testbench.
Yes a to drive a reset is nothing special so an general io vc is probably smart. |
added reset output to use when the axi_stream_master is generating the reset.
changed wait until idle construction to use a notify from the bus process so that we are sure all transactions (including the reset) have been handled.
added checking of reset input on places where it is required so that a reset is handled properly in the master.
added axi_stream_reset to vci.
added reset input to axi_stream_slave.vhd.
added basic test for reset functionality.