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

bug in LWA, SWA and snooping hit #95

Open
hurjaewon opened this issue Oct 29, 2019 · 3 comments
Open

bug in LWA, SWA and snooping hit #95

hurjaewon opened this issue Oct 29, 2019 · 3 comments

Comments

@hurjaewon
Copy link

After LWA instruction, if a snoop_en_i signal and snoop to same address are inserted at some specific time, the reservation does not get cancelled.
According to OpenRISC isa, they said snooping hit should break the reservation.
Isn't it bug?

Here's waveform
Screenshot from 2019-10-29 20-21-08

@stffrdhrn
Copy link
Member

Hi @hurjaewon ,

Thanks for the report, it sounds like a bug. Do you have any code/setup to reproduce it? If so it would be helpful.

@stffrdhrn
Copy link
Member

In your trace can you provide signals for:

  • mor1kx_lsu_cappuccino . dbus_adr_o
  • mor1kx_lsu_cappuccino . dbus_ack_i
  • mor1kx_lsu_cappuccino . snoop_valid

Do you know where the second snoop to 0x2000 is coming from? Are you sure it is external?

@JaewonHur
Copy link
Contributor

Sorry, I missed to reply.
Actually I'm trying with one mor1kx cpu and manually making inputs.
So there is only one snoop as in the figure and other signals are like this.

I was testing with DATA_CACHE=ENABLED, and all the codes are as same as the master branch.
Thank you.

Screenshot from 2019-12-02 11-03-09

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