[Libre-soc-dev] daily kan-ban update 21nov2021

lkcl luke.leighton at gmail.com
Sun Nov 21 14:45:17 GMT 2021

yesterday (and today morning)

* guided sadoon at 3am through install-hdl-dev-repos
   as best i could.
* checked core.py video (all good) which took 8 hours
   to upload
* put some guidance notes for Jacob on core.py
   and why FU proliferation is problematic (has
   to be kept under control)
* thought through hazard bitvectors and realised
   ironically actually they are 100% exactly a FU-Regs
   Dependency Matrix, down to the last detail
* realised that fixing the combinatorial loop requires
   initiating bitvector read at *issue* time and allowing
   matching outstanding register writes to hook into
   GO_READ (fu.go.rd_i), reason being: read requests
   start one cycle *after* issue, and that is exactly
   when the bitvector read results will arrive
* put some more sane (not 1am) notes into pinmux
   bug#50 for Andrey
* created notes for new VLSI coriolis2 NLnet Grant,
   meeting with NLnet Tuesday, to discuss milestones
* found Mitch's CVC, invited him to participate
* tried reaching out to efabless again, suspect they
   are just madly busy with the MPW3 deadline
* thinking through tasks for Dmitry which can be both
   useful and what he will enjoy (c)


More information about the Libre-soc-dev mailing list