This commit adds support for real-valued parameters in blackboxes. Additionally,
parameters now retain their types are no longer all encoded as strings.
There is a caveat with this implementation due to my limited knowledge of yosys,
more specifically to how yosys encodes bitwidths of parameter values. The example
below can motivate the implementation choice I took. Suppose a verilog component
is declared with the following parameters:
parameter signed [26:0] test_signed;
parameter [26:0] test_unsigned;
parameter signed [40:0] test_signed_large;
If you instantiate it as follows:
defparam <inst_name> .test_signed = 49;
defparam <inst_name> .test_unsigned = 40'd35;
defparam <inst_name> .test_signed_large = 40'd12;
If you peek in the RTLIL::Const structure corresponding to these params, you
realize that parameter "test_signed" is being considered as a 32-bit value
since it's declared as "49" without a width specifier, even though the parameter
is defined to have a maximum width of 27 bits.
A similar issue occurs for parameter "test_unsigned" where it is supposed to take
a maximum bit width of 27 bits, but if the user supplies a 40-bit value as above,
then yosys considers the value to be 40 bits.
I suppose this is due to the type being defined by the RHS rather than the definition.
Regardless of this, I emit the same widths as what the user specifies on the RHS when
generating firrtl IR.
Previous blackbox components were just emitted with their interface ports,
but their generic parameters were never emitted and it was therefore
impossible to customize them.
This commit adds support for blackbox generic parameters, though support
is only provided for INTEGER and STRING parameters. Other types of
parameters such as DOUBLEs, ..., would result in undefined behavior here.
This allows the emission of custom extmodule instances such as the following:
extmodule fourteennm_lcell_comb_<instName>:
input cin: UInt<1>
output combout: UInt<1>
output cout: UInt<1>
input dataa: UInt<1>
input datab: UInt<1>
input datac: UInt<1>
input datad: UInt<1>
input datae: UInt<1>
input dataf: UInt<1>
input datag: UInt<1>
input datah: UInt<1>
input sharein: UInt<1>
output shareout: UInt<1>
output sumout: UInt<1>
defname = fourteennm_lcell_comb
parameter extended_lut = "off"
parameter lut_mask = "b0001001000010010000100100001001000010010000100100001001000010010"
parameter shared_arith = "off"
The $div and $mod cells use truncating division semantics (rounding
towards 0), as defined by e.g. Verilog. Another rounding mode, flooring
(rounding towards negative infinity), can be used in e.g. VHDL. The
new $modfloor cell provides this flooring modulo (also known as "remainder"
in several languages, but this name is ambiguous).
This commit also fixes the handling of $mod in opt_expr, which was
previously optimized as if it was $modfloor.
The current firrtl backend emits blackboxes as standard modules
with an empty body, but this causes the firrtl compiler to
optimize out entire circuits due to the absence of any drivers.
Yosys already tags blackboxes with a (*blackbox*) attribute, so this
commit just propagates this change to firrtl's syntax for blackboxes.
The $paramod name mangling is not invertible (the \ character, which
separates the module name from the parameters, is valid in the module
name itself), which does not stop people from trying to invert it.
This commit makes it easy to invert the name mangling by storing
the original name explicitly, and fixes the firrtl backend to use
the newly introduced attribute.
In the past I was calling the ILANG_BACKEND::dump_const() to dump
values to an output stream. When these values were strings, the
function used to add quotes around them. The firrtl compiler, in turn,
escaped these quotes and the result was double-quoted strings which
were hard to read.
However I'm now calling design_entity->get_src_attribute() directly
and there is no additional quote being put around it, so we can
safely remove the unnecessary call to str.erase() here.
Yosys puts quotes around the string that represents the fileinfo whereas
firrtl does not. So when firrtl sees quotes, it escapes them with an extra
backslash which makes it hard to read afterwards.
Use FIRRTL spec vlaues for definition of FIRRTL widths.
Added support for '$pos`, `$pow` and `$xnor` cells.
Enable tests/simple/operators.v since all operators tested there are now supported.
Disable FIRRTL tests of tests/simple/{defvalue.sv,implicit_ports.v,wandwor.v} since they currently generate FIRRTL compilation errors.
Don't emit subfield assignments: bits(x, y, z) <= ... - but instead, add them to the reverse-wire-map where they'll be treated at the end of the module.
Enable tests which were disabled due to incorrect treatment of subfields.
Assume the `$firrtl2verilog` variable contains any additional switches to control verilog generation (i.e. `--no-dedup -X mverilog`)
Issue warning messages for missing parameterized modules and attempts to set initial values.
Replace simple "if (cell-type)" with "else if" chain.
Fix FIRRTL shift handling.
Add support for parameterized modules, $shift, $shiftx.
Handle default output file.
Deal with no top module.
Automatically run pmuxtree pass.
Allow EXTRA_FLAGS and SEED parameters to be set in the environment for tests/tools/autotest.mk.
Support FIRRTL regression testing in tests/tools/autotest.sh
Add xfirrtl files to test directories to exclude files from FIRRTL regression tests that are known to fail.
Don't pad logical operands to one bit.
Use operand width and signedness in $reduce_bool.
Shift amounts are unsigned and shouldn't be padded.
Group "is invalid" with the wire declaration, not its use (otherwise it is incorrectly wired to 0).
Marcelina Kościelnicka
Sahand Kashani
whitequark
Xiretza
Alberto Gonzalez
Eddie Hung
Miodrag Milanovic
Clifford Wolf
Jim Lawson
Kristoffer Ellersgaard Koch