t27 · tri vectors debt
fpga_gf16_accel.json 27 0 prose-only (no data at all)
fpga_hir.json 26 0 prose-only (no data at all)
fpga_hw_types.json 22 0 prose-only (no data at all)
fpga_linker.json 17 0 prose-only (no data at all)
fpga_mac_vectors.json 21 21 executed
fpga_memory.json 15 0 prose-only (no data at all)
fpga_partition.json 12 0 prose-only (no data at all)
fpga_placement.json 14 0 prose-only (no data at all)
fpga_power.json 15 0 prose-only (no data at all)
fpga_power_analysis.json 9 9 debt (has data, no runner)
fpga_router.json 19 0 prose-only (no data at all)
fpga_simulator.json 13 0 prose-only (no data at all)
fpga_spi.json 8 8 executed
fpga_ternary_isa.json 29 0 prose-only (no data at all)
fpga_testbench_mac_tb.json 10 10 debt (has data, no runner)
fpga_testbench_top_tb.json 4 4 debt (has data, no runner)
fpga_testbench_uart_tb.json 7 7 debt (has data, no runner)
fpga_timing.json 24 0 prose-only (no data at all)
fpga_top_level.json 7 7 debt (has data, no runner)
fpga_uart.json 7 7 debt (has data, no runner)
fpga_vcd_conformance_compare.json 16 16 debt (has data, no runner)
fpga_vcd_trace.json 18 0 prose-only (no data at all)
2 executed, 8 debt, 24 prose-only (34 files); 512 cases, 100 carrying data (19%).
PROSE-ONLY is the majority and the important number: those cases carry
an id and a sentence, no inputs and no expected values, so no runner
can ever execute them as written -- they are documentation shaped like
tests. DEBT is executable in principle and blocked by numbered defects
(#2410 slices, #2413 test emission). Neither counts as coverage.
t27 $