cocotb_benches.dau_fixed_point_exp_tb¶

The rendered exp tile against the model it was rendered from.

Zero tolerance: the HDL mirrors ExpDatapath.evaluate step for step, so any difference is a defect in one of them. Deviation from correctly rounded exp is a separate, measured question answered in test_fixed_point_exp.

Attributes¶

`_MASK32`
`_GENERATOR`
`_POINT`

Functions¶

`_signed32`(value) → int
`matches_the_model_on_directed_inputs`(dut)
`matches_the_model_across_the_domain`(dut)
`_start`(dut) → None
`_run`(dut, rows, *[, ready_pattern])
`_send`(dut, words) → None
`_collect`(dut, *, expect[, ready_pattern])	The proven quad-row collector: drive ready, sample in ReadOnly, and

Module Contents¶

cocotb_benches.dau_fixed_point_exp_tb._MASK32 = 4294967295¶

cocotb_benches.dau_fixed_point_exp_tb._GENERATOR¶

cocotb_benches.dau_fixed_point_exp_tb._POINT¶

cocotb_benches.dau_fixed_point_exp_tb._signed32(value: int) → int¶

async cocotb_benches.dau_fixed_point_exp_tb.matches_the_model_on_directed_inputs(dut)¶

async cocotb_benches.dau_fixed_point_exp_tb.matches_the_model_across_the_domain(dut)¶

async cocotb_benches.dau_fixed_point_exp_tb._start(dut) → None¶

async cocotb_benches.dau_fixed_point_exp_tb._run(dut, rows, *, ready_pattern=(1,))¶

async cocotb_benches.dau_fixed_point_exp_tb._send(dut, words) → None¶

async cocotb_benches.dau_fixed_point_exp_tb._collect(dut, *, expect: int, ready_pattern=(1,))¶: The proven quad-row collector: drive ready, sample in ReadOnly, and only count a beat that actually fired. Sampling without checking the handshake is what made this harness report zeros against a tile the waveform showed was correct.