Has anyone tested any of the OpenRAM macros on MPW2 yet? I have some
sram_1rw1r_32_256_8_sky130
that on a very brief test aren't responding (but I certainly can't rule out a problem in my test setup or integration at this point, so don't treat it as a writeoff).
gatecat
02/13/2023, 5:12 PM
ah, actually, I'm curious, does OpenRAM have a minimum clock? I see some life out of it when running at 10MHz, whereas at 1kHz (ish) I just get zeros out
gatecat
02/13/2023, 5:19 PM
ahhh, pebcak... I missed that the data is invalid between the posedge and negedge. checking read data at the negedge now it works fine at 1kHz too
m
Matthew Guthaus
02/13/2023, 5:35 PM
We are still doing bring-up on our test chip. We tested essentially that same macro pre-MPW and it was working up to ~12MHz.
Matthew Guthaus
02/13/2023, 5:36 PM
(It was limited by IO delays)
👍 2
Matthew Guthaus
02/13/2023, 5:36 PM
Yes, data should become valid some time after the negative edge. The first half of the cycle is precharge and then read occurs in the second half.
Matthew Guthaus
02/13/2023, 5:40 PM
Actually, that macro is the exact one we tested in silicon. Not the updated one.
Linen is a search-engine friendly community platform. We offer integrations with existing Slack/Discord communities and make those conversations Google-searchable.