feat: add FPGA-Neural benchmark tooling + same-tier ECP5 P2/P4/P8 results

tools/fpga_benchmark.py: parametric Yosys + nextpnr-ecp5 benchmark harness for the LFE5U-45F-8BG381 (speed grade -8, 80 MHz target), sweeping PARALLEL over the neuron layer and parsing Fmax/LUT4/DFF/DSP utilization out of the nextpnr report into JSON/CSV. synth/ecp5/p2, p4, p8: real synthesis+PnR results backing the same-price-tier FPGA comparison (P2: 87.88 MHz PASS, P4: 75.01 MHz FAIL, P8: 147.62 MHz PASS -- non-monotonic, dominated by placement noise since the whole design uses <2% of the device's LUT4 fabric at every setting, and P8 notably maps to 0 DSP blocks vs 8/16 for P2/P4). synth/ecp5/top.v: benchmark harness top-level, reworked to generate deterministic non-constant X/weights/bias via `keep`-attributed generate blocks so Yosys can't constant-fold the datapath away. Also adds .gitignore for Python's __pycache__/*.pyc.
2026-09-02 19:48:03 +02:00
parent 233d6ff7fb
commit cfd5e98a0e
28 changed files with 230527 additions and 49 deletions
@@ -0,0 +1,4 @@
+target_id,fpga,family,parallel,n_inputs,n_neurons,status,fmax_mhz,period_ns,slack_ns,lut,ff,dsp,lut_pct,ff_pct,dsp_pct,mac_per_cycle,groups_per_neuron,total_mac,cycles_per_layer,cost_eur,mac_per_second,mac_per_euro,build_time_s,error
+ecp5_45f,LFE5U-45F-8BG381,ECP5,8,256,4,FAIL,,,,,,,,,,32,32,32,32,35.0,,,,nextpnr placement/routing failed
+ecp5_45f,LFE5U-45F-8BG381,ECP5,4,256,4,FAIL,,,,,,,,,,16,64,16,64,35.0,,,,nextpnr placement/routing failed
+ecp5_45f,LFE5U-45F-8BG381,ECP5,2,256,4,PASS,,,,,,,,,,8,128,8,128,35.0,,,1.504363291998743,
@@ -0,0 +1,83 @@
+[
+  {
+    "target_id": "ecp5_45f",
+    "fpga": "LFE5U-45F-8BG381",
+    "family": "ECP5",
+    "parallel": 8,
+    "n_inputs": 256,
+    "n_neurons": 4,
+    "status": "FAIL",
+    "fmax_mhz": null,
+    "period_ns": null,
+    "slack_ns": null,
+    "lut": null,
+    "ff": null,
+    "dsp": null,
+    "lut_pct": null,
+    "ff_pct": null,
+    "dsp_pct": null,
+    "mac_per_cycle": 32,
+    "groups_per_neuron": 32,
+    "total_mac": 32,
+    "cycles_per_layer": 32,
+    "cost_eur": 35.0,
+    "mac_per_second": null,
+    "mac_per_euro": null,
+    "build_time_s": null,
+    "error": "nextpnr placement/routing failed"
+  },
+  {
+    "target_id": "ecp5_45f",
+    "fpga": "LFE5U-45F-8BG381",
+    "family": "ECP5",
+    "parallel": 4,
+    "n_inputs": 256,
+    "n_neurons": 4,
+    "status": "FAIL",
+    "fmax_mhz": null,
+    "period_ns": null,
+    "slack_ns": null,
+    "lut": null,
+    "ff": null,
+    "dsp": null,
+    "lut_pct": null,
+    "ff_pct": null,
+    "dsp_pct": null,
+    "mac_per_cycle": 16,
+    "groups_per_neuron": 64,
+    "total_mac": 16,
+    "cycles_per_layer": 64,
+    "cost_eur": 35.0,
+    "mac_per_second": null,
+    "mac_per_euro": null,
+    "build_time_s": null,
+    "error": "nextpnr placement/routing failed"
+  },
+  {
+    "target_id": "ecp5_45f",
+    "fpga": "LFE5U-45F-8BG381",
+    "family": "ECP5",
+    "parallel": 2,
+    "n_inputs": 256,
+    "n_neurons": 4,
+    "status": "PASS",
+    "fmax_mhz": null,
+    "period_ns": null,
+    "slack_ns": null,
+    "lut": null,
+    "ff": null,
+    "dsp": null,
+    "lut_pct": null,
+    "ff_pct": null,
+    "dsp_pct": null,
+    "mac_per_cycle": 8,
+    "groups_per_neuron": 128,
+    "total_mac": 8,
+    "cycles_per_layer": 128,
+    "cost_eur": 35.0,
+    "mac_per_second": null,
+    "mac_per_euro": null,
+    "build_time_s": 1.504363291998743,
+    "error": null
+  }
+]