I put it under dff2dffe, but get no conversions...
-
-
Replying to @nachiketkapre @oe1cxw
Some progress, (1) Put shregmap before cell_map.v pass, (2) clkpol and enpol set to "pos"
1 reply 0 retweets 0 likes -
Replying to @nachiketkapre @oe1cxw
I see "Converting .. to a shift register" messages, but later on get an error "GP_SHREG" not part of design during Analyze Hierarchy
1 reply 0 retweets 0 likes -
Replying to @nachiketkapre @oe1cxw
Ahh GP_SHREG is only a GreenPak4 primitive.. Need to find the Xilinx version
1 reply 0 retweets 1 like -
Replying to @nachiketkapre @oe1cxw
Almost there -- the SRLs are getting thrown out later -- think I need to configure A0-A3 inputs with constants..
1 reply 0 retweets 0 likes -
Replying to @nachiketkapre @oe1cxw
Not sure what's the mechanism to initialize ports with 0 or 1 constants...
1 reply 0 retweets 0 likes -
Replying to @nachiketkapre @oe1cxw
RTLIL.SigSpec(false/true)?
1 reply 0 retweets 0 likes -
Replying to @nachiketkapre @oe1cxw
That's not helping -- opt_clean is somehow detecting the SRLs as "unused" bocks
1 reply 0 retweets 0 likes -
Replying to @nachiketkapre
are the SRL outputs connected to a cone that contributes to a module primary output?
2 replies 0 retweets 0 likes -
Replying to @oe1cxw
I was relying on "first_cell->setPort(q_port, last_cell->getPort(q_port));" in process_chain() to wire SRL outputs for me
2 replies 0 retweets 0 likes
That is only for the Q port of the internal cell. The one that is removed after fixup() has been run. (If fixup() ret false).
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.