~ update interactive script for chip_io.v for recent openlane
~ update config.tcl for recent openlane
~ add a verilog stub for sky130_fd_io__top_xres4v2 as
the io verilog models are not readable by yosys
- reimplement using a sparecell
- reimplement using newest open_pdks
!important using openlane pre odb with some local patches which
most if not all are merged in the current head of openlane however
still takes effort to update the interactive script to be latest
openlane compatible
!important override abstract lef generated by openlane. openlane
generates lef and mag that contain def BLOCKAGE layers that cause
congestions during top level routing
- remove unused port in chip_io
- move the rest of chip_io power ports to the USE_POWER_PINS guard
- add caravel_power_routing cell guarded by TOP_ROUTING ifdef
~ update the following views:
def
mag
verilog
spef(all corners)
+ add the ability to override the interactive script filename
+ add the ability to run openlane regression using regression.config
file
~ change GRT ADJUSTMENT values
~ change pointers to some files for workarounds
!important the interactive script still needs updates
!important this was done using old openlane v0.22 and its matching
pdk
!important known workarounds:
- a custom techlef is used where large metal spacing rules are the
only ones present to avoid violations by the router
- some odd behaviour happening when a macro has a lef view
with a non zero origin. so the power routing cell is (temporarily)
modified to have a zero origin and its placement has been shifted
which doesn't match the power routing mag.
- the old openlane doesn't generate multi spef corners. they
are generated using timing-scripts repo
includes RTL change inside the padframe definition to remove one
previously unnoticed hard-wired connection between VDDIO and a
3.3V domain digital input pin.
* Effectively reverted the change to add spare logic blocks near each
of the GPIO control blocks by changing the definition of
NUM_SPARE_BLOCKS to 4 (the original number of spare logic blocks)
for both caravel and caravan top level RTL verilog modules.
* Apply automatic changes to Manifest and README.rst
Co-authored-by: RTimothyEdwards <RTimothyEdwards@users.noreply.github.com>
- rtl updated
~ add one column to the right to pass placement congestion
~ density adjusted (probably has no effect)
+ manually add isosubstrate layer in mag and gds from older iterations
!important still need to run dynamic simulations
!important depends on some updates to openlane
!important need to be able to recreate using newer openlane versions
* Added enough spare logic blocks to have the existing four above
the processor, plus one each per GPIO (38 for caravel, 27 for
caravan).
* Apply automatic changes to Manifest and README.rst
Co-authored-by: RTimothyEdwards <RTimothyEdwards@users.noreply.github.com>
Co-authored-by: Jeff DiCorpo <42048757+jeffdi@users.noreply.github.com>
* Removed the SRAM read-only interface by wrapping all related code
in an ifdef for "USE_SRAM_RO_INTERFACE", which is undefined.
* Apply automatic changes to Manifest and README.rst
Co-authored-by: RTimothyEdwards <RTimothyEdwards@users.noreply.github.com>
Co-authored-by: Jeff DiCorpo <42048757+jeffdi@users.noreply.github.com>
to caravel_netlists.v when attempting to determine if the
verification testbenches could be run from caravel referencing
caravel_mgmt_soc_litex instead of the other way around. This file
has been reverted back to its original form.
This reverts commit 577cc12fe0.
Reverting the change from issue #16. After some discussion, it has
been decided that it is up to the user to implement the pull-up and
pull-down modes correctly by setting the output enable and driving
the output to the appropriate value. Note that this should be well
documented, if by nothing else than a validation testbench that
excercises a user pull-up and pull-down input mode.
zero level outputs when the user project area is powered down.
That allows the synthesis tools to buffer these outputs. The
protection from floating inputs is left as-is, but all logic that
was unnecessary to be specified by gate instances has been changed
to RTL. This leaves only a handful of signals (logic analyzer input,
user IRQ, and wishbone data out and acknowledge out) to be handled
by explicit logic gate instances.
the left hand side of caravan from gpio_defaults_block_14 to
gpio_defaults_block_25. Otherwise, the script that generates the
custom user configuration won't be able to change the defaults
for GPIO 25.
- high level changes:
* add larger buffers on output ports
* add buffers on input ports
* adjust sdc file increasing output load and setting a high transition
- detailed changes:
* add interactive script for openlane where the order of events is a bit shuffled
- to add obstruction before pdn
- to manually insert buffers on some ports
- to manually remove buffers inserted by synthesis on for example serial_clock_out
* change openlane config adding extra row and columns to increase the space and fit the
added buffers
* change config to enable buffering
* increase density for better placement?
* change the cell exclude list. some excluded cells didn't make sense
* ef decap cells break dynamic sims?
* add custom pdn script for to duplicate the old pdn
- misc changes:
* fix openlane makefile to properly detect interactive script
!important still need to run dynamic simulations
!important depends on some updates to openlane
Corrected the pad constant connections to all be in the correct domain
(1.8V or 3.3V). Created a new "constant_block" module that generates
a single constant 1 and 0 value in the 1.8V domain, and used 7 of these
in the chip_io (and chip_io_alt) modules to create the 1.8V domain
constant signals for the seven pads belonging to the management (clock,
reset, flash SPI, and management GPIO).
inputs that are permanently tied low or high come from either
the local "TIE" pad connections (if they are in the 3.3V
domain) or from a constant one wire in the 1.8V domain that
is generated in the gpio_control_block module and exported
to the chip_io (or chip_io_alt) module.
from the data and replace them with a single flop clocked on the
negative edge of the serial clock. This will completely avoid hold
violations by ensuring that the block's output data bit does not
change anywhere near the clock rising edge, so clocks do not have
to be tightly aligned among all of the GPIO blocks.
does not pass through an inverter, so that the input can remain
unconnected. Rewired the existing implementation to use an
alternative gate that has an inverting input so that the
user_gpio_out signal can be left undriven when the GPIOs are in
the management enable state. This is a simple logic refactoring
and does not change the logic function. The manual rewiring has
been confirmed by LVS, but at least one GL simulation should be
run to confirm that the logic function remains the same as before.
* Corrects four signals which were missing from the caravan top level
(management output and output enable to GPIO 0 and 1---these errors
would have prevented the houskeeping SPI from working on caravel).
Corrected RTL verilog (source of the error), GL verilog, and layout.
* Apply automatic changes to Manifest and README.rst
Co-authored-by: RTimothyEdwards <RTimothyEdwards@users.noreply.github.com>
* (1) Modified the .magicrc file to set a default for PDK if not set in the
environment. (2) Fixed the user ID programming layout to not leave holes
behind when the script moves the vias around (similar to the handling of
the GPIO defaults block). (3) Added substrate isolation to gpio_control_block
and fixed the path references to the standard cells. (4) Fixed the four
missing routes on the Caravan top level. (5) Reinstated the large rendered
labels for the pads on both caravel and caravan. (6) Corrected the top
level gate-level netlist for caravan to add the missing pins to the
management core wrapper. (7) Did the same for the caravan top level RTL.
(8) Created scripts to run full LVS including extracting the management
core wrapper and reading all gate-level verilog submodules. (9) Moved all
of the LVS scripts to the scripts directory.
* Apply automatic changes to Manifest and README.rst
* Made the changes from pull request #73 as they did not get merged
successfully, and if merged now they will generate conflicts with
this pull request in scripts/set_user_id.py. So it's easier to
just manually add them to this pull request.
* Apply automatic changes to Manifest and README.rst
Co-authored-by: RTimothyEdwards <RTimothyEdwards@users.noreply.github.com>
* - update caravan configuration
* change correct gpio_default blocks instance names
* implement powerrouting work around
* place spare logic blocks
* keep the history of runs and sym link last/final one
* - update caravan mag, def and gl views
- add fake power routing cell in the rtl as a placeholder for openlane
to prevent routing on that cell
- fix some syntax errors in chip_io_alt
- match simple_por lef pins with mag by regenerating it
* Apply automatic changes to Manifest and README.rst
* add caravan power routing lef
* - update mag and def view of caravan
- add_macro_placement for fake cell
* Added back the power routing to Caravan, fixed DRC errors, ran LVS,
corrected placement of isolated substrate regions, and replaced the
signal routing for the analog lines. Be aware that merging with
main may cause issues with the mgmt_protect.mag file and its
subcells mprj2_logic_high.mag and mprj_logic_high.mag. It may be
worth cherry-picking the files to merge and exclude those layouts.
Co-authored-by: kareem <kareem.farid@efabless.com>
Co-authored-by: kareefardi <kareefardi@users.noreply.github.com>
Co-authored-by: Jeff DiCorpo <42048757+jeffdi@users.noreply.github.com>
kareem
M0stafaRady
Mohamed Shalan
Mohamed Hosni
mo-hosni
M0stafaRady
Tim Edwards
Jeff DiCorpo
Marwan Abbas
jeffdi
Kareem Farid