LOFMPL: An Open-Source Logic Optimization Framework with Maximum Fanout-Free Cone (MFFC)-based Hypergraph partitioning and Reinforcement Learning

Prerequisites

System packages (Ubuntu 22.04 / Debian / Arch):

g++  cmake  make  git  python3  yosys

Install on Ubuntu/Debian: sudo apt install g++ cmake make git python3 yosys

Micromamba (conda-compatible package manager, required for the RL Python environment):

"${SHELL}" <(curl -L micro.mamba.pm/install.sh)

Quick Build

git clone --recursive https://github.com/YOUR_USERNAME/LOFMPL.git
cd LOFMPL
make all        # builds abc_p, LSOracle_p, abc_netlist.so, abc_py

Individual targets: make abc, make lsoracle, make abc_netlist, make abc_py

For Python RL environment setup and pipeline usage, see SETUP.md.

---

As circuit complexity increases, traditional reinforcement learning (RL) approaches face significant challenges in effectively exploring logic optimization sequences within large-scale Boolean networks. These methods often suffer from long runtime overhead and suboptimal optimization results.

LOFMPL addresses these challenges by introducing an innovative logic optimization framework that combines Maximum Fanout-Free Cone (MFFC)-based Hypergraph partitioning with reinforcement learning. LOFMPL leverages Yosys for Verilog parsing and combinational logic extraction.

The core of LOFMPL is a novel two-stage MFFC-based hypergraph partitioning technique. This technique partitions the circuit into highly independent subnetworks, enabling efficient exploration by a parallel RL-based design space exploration engine implemented using C++ APIs. This approach reduces runtime while enhancing the quality of logic optimization.

Key Features

• Enhanced Optimization: Achieves superior optimization results within the same runtime constraints compared to state-of-the-art methods.
• Broad Benchmarking: Validated against more than 150 benchmarks compared to other ML-based and greedy methods.
• Significant Improvements:
  • Partition Efficiency: Outperform KaHypar and MFFCKaHypar by 33% and 28% on node-level-product without significant runtime overhead.
  • Logic Optimization Task: Average node-level-product improvement of 13% over RLG, 3% over ESE, 14% over Boils, and 7% over DRiLLS.
  • ASIC Technology Mapping Task: Average Area-Delay-Product (ADP) improvement of 23% over LSOracle, 9% over Boils, and 5% over DRiLLS.
  • Comprehensive Benchmarking: Across all 150+ benchmarks, LOFMPL demonstrates average improvements of 9%, 6%, and 7% compared to ABC 'resyn2', BOiLS, and DRiLLS respectively.

LOFMPL is an open-source project designed to advance the field of logic optimization, offering a robust, scalable solution for large-scale circuit design challenges.

If you have any questions, please feel free to contact my email for inquiries: kxzhu21@m.fudan.edu.cn

Build

To build and configure LOFMPL, follow the steps below:

1. Clone the Repository

First, clone the LOFMPL repository and initialize the required submodules:

git clone https://github.com/kxzhu0505/LOFMPL.git
cd LOFMPL
git submodule update --init --recursive

2. Build ABC and LSOracle

LOFMPL relies on ABC and LSOracle for circuit partitioning. You need to build these tools following their respective instructions.

ABC_P:

• Navigate to the ABC submodule directory:

  cd abc_p

• Build ABC as per the instructions in the README.md of the ABC repository.

LSOracle_p:

• Navigate to the LSOracle submodule directory:

  cd LSOracle_p

• Follow the build instructions in the README.md of the LSOracle repository.

3. Build rl_logic_synthesis for Parallel Optimization

LOFMPL uses the rl_logic_synthesis tool for reinforcement learning-based optimization. Build it by following the steps below:

• Navigate to the rl_logic_synthesis directory:

  cd rl_logic_synthesis

• Follow the build and setup instructions provided in the README.md of the rl_logic_synthesis repository.

4. Build abc_netlist.so

• Navigate to the abc_netlist directory:

  cd abc_netlist
  g++ -shared -o abc_netlist.so abc_netlist.cc

Usage

Once the build process is complete, follow these steps to perform partitioning, optimization, and evaluation of circuits.

To perform the optimization, use the run_top.sh script:

./src/run_top.sh

This step divides the input circuit using tools such as ABC and LSOracle.

---

Manual Partition and Reassemble (Noop Pipeline)

Use these steps to partition a design and glue it back without RL optimization. This is useful for verifying the partition round-trip or as a starting point before applying RL.

1. Create a work directory with bench_info.yaml

file(s):
  - /abs/path/to/dep.v
  - /abs/path/to/top.v
top: your_top_module
type: sequential   # or combinational

2. Partition — yosys flattens the design, abc_p pif partitions it, LSOracle further splits large partitions with --noop_dir

cd /your/work/dir
python3 src/partition.py \
  -work_dir . \
  -yosys_exe /usr/bin/yosys \
  -abc_exe   abc_p/abc \
  -abc_p_exe abc_p/abc \
  -lsoracle_exe LSOracle_p/build/core/lsoracle \
  -part_size 400    # tune: smaller = more partitions

Output: src/part_N.v, src/part_N.ports, src/top.inputs, src/top.outputs

3. Reassemble combinational top

python3 src/gen_top.py -dir src -module_name netlist

Output: src/netlist.v

4. Flatten the assembled netlist

yosys -Q -T -p \
  'read_verilog src/*.v;
   hierarchy -top netlist;
   flatten; techmap; opt -purge;
   write_verilog -noattr netlist.v'

Output: netlist.v (flat, in work dir)

5. Re-insert registers — get final sequential output

This step reads the original RTL (with DFFs), verifies equivalence via ABC CEC, and substitutes the optimized combinational core, producing a sequential output.v.

yosys -Q -T \
  -m abc_netlist/abc_netlist.so \
  -p 'read_verilog -nolatches /abs/path/dep.v;
      read_verilog -nolatches /abs/path/top.v;
      hierarchy -check -top your_top_module;
      flatten; proc; opt -purge; memory; opt -purge;
      fsm; opt -purge; techmap; opt -purge;
      abc_netlist -netlist netlist.v -script "+cec netlist.v; quit";
      opt -purge; write_verilog -noattr src/output.v'

Output: src/output.v — sequential, same interface as original RTL

Note: The --noop_dir partitions are intentionally combinational — they represent the logic between pipeline registers. Register re-insertion happens in step 5 via the abc_netlist yosys plugin. For sequential designs, CEC verifies equivalence of the combinational abstraction (DFFs exposed as PI/PO). A noop round-trip produces a bit-exact equivalent of the original.

---

RL Optimization — src/rl_opt.py

rl_opt.py is the main tool for RL-based optimization. It takes a folder of partition .v files, trains PPO on each, and writes optimized .v files to an output folder. Simulation, verification, and metrics are left to you.

python3 src/rl_opt.py \
  --input   path/to/partitions/ \
  --output  path/to/optimized/ \
  --steps   5000 \
  --workers 4

Option	Default	Description
--input	required	folder with input partition .v files
--output	required	folder for optimized .v files (created if missing)
--steps	5000	PPO training steps per partition
--workers	4	parallel training jobs (each spawns ABC subprocesses — don't set too high)
--min-nodes	5	skip partitions smaller than N AND-nodes (trivial)
--work-dir	<output>/.rl_work	where PPO logs and monitor CSVs are stored

Resume-safe: re-running the command skips any partition whose 0.monitor.csv already exists in --work-dir. Interrupt and restart freely.

Output: one optimized .v per input partition. Trivial partitions (below --min-nodes) are copied unchanged. Partitions with no monitor (training failed) are also copied unchanged.

Example — ifft8, 24 partitions, 5000 steps:

Partition                                 Before   After  Saved%
------------------------------------------------------------
network_5_part_0_noop                        677      74   89.1%
network_6_part_4_noop                        671      63   90.6%
network_6_part_8_noop                        672      66   90.2%
network_0                                    593     360   39.3%
...
TOTAL                                      11098    6023   45.7%

Reassembly (optional — if you need a single flat netlist):

python3 src/gen_top.py -dir path/to/optimized/ -module_name netlist
yosys -Q -T -p "read_verilog path/to/optimized/*.v; \
  hierarchy -top netlist; flatten; techmap; opt -purge; \
  write_verilog -noattr flat_netlist.v"

Observed results on ifft8 (8-point IFFT, 24 partitions, 5000 steps):

• AND-node reduction: 45.7% (11,098 → 6,005 nodes)
• NMSE vs original RTL: −∞ dB (bit-exact)

---

Experiment Results For All 150+ benchmarks on ASIC Technology Mapping

Benchmarks		ABC			LSOracle			BoiLs			DriLLs			LOFMPL
	Area(um2)	Delay(ps)	ADP	Area(um2)	Delay(ps)	ADP	Area(um2)	Delay(ps)	ADP	Area(um2)	Delay(ps)	ADP	Area(um2)	Delay(ps)	ADP
/vtr/arm_core	156318.2	1034.19	161662747.9	151004.5	1056.51	159537767.5	152004.9	1041.54	158319148	161500.8	1031.03	166512130	146799.1	1022.63	150121140
/OPDB/sparc_ifu	82344.63	1418.87	116836326.3	82610.32	1623.1	134084814.1	80266.22	1366.24	109662924	77454.7	1378.32	106757366	77491.85	1365.92	105847672
/quip/oc_aquarius	74737.33	1319.75	98634586.27	72260.9	1322.08	95534688.89	76614.06	1289.77	98814521	65852.52	1310.53	86301706.2	65919.68	1324.6	87317205.4
/koios/spmv	69036.24	486.8	33606841.71	68554.43	442.71	30349731.48	69442.06	518	35970987	67610.2	441.99	29883032.1	66835.2	427.62	28580068.1
/vtr/mcml	64033.73	2422.24	155105052.5	65656.8	2516.05	165195793.6	63919.26	2408.69	153961679	63803.23	2477.33	158061658	62014.1	2412.85	149630732
/koios/attention_layer	58978.97	672.35	39654511.41	54964.41	655.77	36044012.64	52220.98	675.19	35259084.3	52565	662.03	34799604.5	52056.65	656.01	34149683
/OPDB/l15_wrap	54529.61	694.74	37883899.31	54230.35	636.12	34497012.02	52146.39	705.92	36811179.3	52314.85	658.1	34428400.7	51671.43	659.52	34078342.5
/OPDB/lsu	47143.15	517.17	24381020.75	52444.45	650.54	34117211.54	15886.88	564.46	8967507.15	15887.74	491.67	7811524.24	15565.55	476.73	7420564.35
/koios/bnn	46684.91	766.98	35806393.41	52291.84	703.69	36797244.93	47817.84	779.37	37267786.7	48742.38	750.25	36568972.7	47142.9	672.42	31699827.8
/vtr/LU8PEEng	30131.14	4459.65	134374356.6	32038.3	4468.85	143174337.3	29928.95	4324.69	129433425	30648.11	4370.53	133948472	29466.34	4179.1	123142781
/OPDB/fpga_bridge_rcv_32	26856.86	489.02	13133539.33	26317.4	492.81	12969475.54	26211.68	437.79	11475209.5	26362.59	455.97	12020551.6	25876.15	499.9	12935485.4
/vtr/sv_chip2_hierarchy_no_mem	17329.58	1039.59	18015661.9	17958.9	1047.64	18814462.22	16616.14	1026.55	17057299.1	16746.82	1004.29	16818665	16128.13	1014.9	16368442.6
/koios/robot_maze	16794.64	368.82	6194200.368	16629.96	331.26	5508841.301	16365.57	368.82	6035948.98	16652.61	342.07	5696356.59	16187.2	321.65	5206611.86
/OPDB/sparc_exu_wrap	15425.38	465.44	7179587.601	14959.43	574.57	8595239.481	15648.44	461.05	7214711.75	15623.78	469.13	7329584.81	15375.83	439.89	6763675.81
/vtr/mkDelayWorker32B	13983.25	428.96	5998257.022	13891.52	434.21	6031835.834	13788.95	433.98	5984127.33	13791.44	482.77	6658093.69	13755.33	424.64	5841061.53
/vtr/sv_chip1_hierarchy_no_mem	10483.11	506.19	5306444.109	10734.42	484.28	5198466.276	10837.47	499.5	5413318.4	10652.57	506.75	5398190.2	10485.94	503.2	5276522.93
/vtr/system	8871.361	639.46	5672880.637	8888.653	636.27	5655583.302	8924.724	610.22	5446045.08	8848.412	633.89	5608920.06	8847.129	619.61	5481769.75
/iwls05/opencores/vga_enh_top	8781.155	322.06	2828058.7	8581.467	319.67	2743237.581	8464.842	346.26	2931036.07	8290.246	327.63	2716133.35	8394.566	324.38	2723029.34
/vtr/bgm	8666.031	860	7452786.803	8759.635	874.37	7659161.849	8689.884	792.87	6889948.37	8410.604	821.83	6912086.81	8225.409	739.68	6084170.53
/koios/top	7834.3	664.41	5205187.54	7767.932	680.06	5282660.014	7699.669	674.58	5194042.51	7736.921	670.08	5184355.75	7678.63	660.27	5069968.88
/quip/oc_video_compression_systems_dct	7757.508	709.1	5500848.656	8111.51	705.42	5722021.398	8171.448	704.5	5756785.41	7897.898	714.94	5646523.52	7766.708	704.86	5474441.53
/quip/oc_video_compression_systems_jpeg	7582.096	638.1	4838135.235	7734.777	631.2	4882191.499	7326.873	646.18	4734478.64	7237.847	643.14	4654949.1	7246.639	657.52	4764810.09
/koios/softmax	7088.096	574.7	4073528.807	7280.45	580.09	4223316.243	7098.39	584.58	4149576.56	7096.013	566.71	4021381.53	7053.017	554.47	3910686.11
/quip/oc_des_des3perf	6363.164	266.97	1698773.864	6877.196	300.73	2068179.262	6084.948	291.12	1771450.16	6377.904	278.75	1777840.82	6089.833	259.03	1577449.35
/OPDB/sparc_ffu_nospu_wrap	5834.26	431.75	2518941.663	5544.701	417.95	2317407.781	5788.551	449.04	2599291.16	5747.13	408.44	2347357.66	5582.93	415.36	2318925.7
/iwls05/opencores/eth_top	5577.2	360.39	2009967.045	5512.756	422.38	2328477.976	5404.441	355.11	1919171.19	5520.659	461.11	2545630.88	5450.368	352.97	1923816.54
/OPDB/tlu_nospu_wrap	5529.523	577.22	3191751.394	6330.681	610.14	3862601.854	5568.248	579.02	3224126.78	5558.187	639.39	3553849.5	5497.258	598.3	3289009.27
/vtr/sv_chip0_hierarchy_no_mem	5308.695	227.37	1207037.873	5300.486	243.28	1289502.228	5261.003	228.25	1200824.01	5405.622	236.44	1278105.35	5288.807	227.96	1205636.53
/vtr/paj_raygentop_hierarchy_no_mem	4919.35	611.17	3006559.291	4892.304	595.75	2914590.315	4711.308	624	2939856.33	4839.787	612.42	2963982.47	4588.545	598.28	2745234.48
/OPDB/chip_bridge	4842.703	471.79	2284738.927	4820.104	480	2313650.001	4828.721	508.88	2457239.51	4875.421	505.32	2463647.58	4754.407	478.48	2274888.51
/vtr/paj_boundtop_hierarchy_no_mem	4815.511	369.97	1781594.777	4719.604	397.61	1876561.837	4670.82	392.17	1831755.3	4671.869	369.47	1726115.55	4602.337	363.68	1673778.02
/vtr/mkPktMerge	3807.713	240.87	917163.7636	3772.415	237.02	894137.6947	3786.397	231.31	875831.436	3970.236	240.47	954722.647	3732.232	225.84	842887.29
/benchmarks/arithmetic/sqrt/top	3696.803	23481.82	86807656.52	3430.426	22816.8	78271347.15	2731.096	14585.37	39834052.1	3035.104	16686.45	50645111.4	2634.271	14199.12	37404325.2
/iwls05/faraday/frisc	3076.351	344.28	1059126.046	3022.623	356.66	1078048.89	2968.153	351.11	1042148.06	3014.663	354.17	1067703.13	2948.543	353.37	1041926.46
/iwls05/faraday/TOP	2839.47	666.02	1891143.512	2964.755	687.55	2038417.64	2838.23	681.02	1932891.57	2861.15	694.87	1988127.31	2302.022	641.04	1475687.92
/benchmarks/arithmetic/log2/top	2325.481	3292.23	7656017.76	2460.2	3336.55	8208580.42	2290.955	3259.56	7467506.56	2051.1	3311.99	6793222.07	1941.327	3331.18	6466909.65
/quip/oc_fpu	2317.622	9028.21	20923980.01	2689.82	9471.38	25476311.61	2443.739	9272.64	22659913.9	2308.247	9109.36	21026655.4	2220.242	8976.81	19930694.1
/iwls05/opencores/fpu	2281.887	9158.31	20898225.15	2577.146	9600.33	24741454.07	2413.894	9194.28	22194016.9	2238.307	9066.59	20293812	2228.188	9027.16	20114214
/vtr/or1200_flat	2070.418	929.24	1923915.498	2130.561	909.13	1936956.739	2100.001	918.18	1928179.03	2113.619	924.61	1954273.12	2086.456	909.91	1898487.45
/quip/oc_aes_core_inv	2027.509	468.93	950759.9632	2158.656	478.1	1032053.652	1971.012	472.3	930908.901	1982.253	478.06	947635.888	1950.469	468.55	913892.081
/iwls05/opencores/wb_conmax_top	1878.998	347.58	653101.9511	1855.043	356.91	662083.239	1854.124	354.62	657509.46	1875.965	337.57	633269.456	1819.978	345.62	629020.677
/iwls05/opencores/des/des	1840.638	265.01	487787.3447	2027.524	282.44	572653.861	1815.808	265.39	481897.222	1808.591	265.33	479873.361	1788.252	259.38	463836.69
/iwls05/opencores/pci_bridge32	1760.812	331.82	584272.6378	1758.99	324.41	570633.784	1760.01	308.79	543473.519	1774.051	308.96	548110.685	1757.342	317.8	558483.275
/quip/oc_des_perf_opt	1732.964	239.08	414317.0817	1881.01	244.64	460170.1692	1760.156	232.21	408725.802	1737.819	240.92	418675.436	1656.594	232.67	385439.764
/benchmarks/arithmetic/multiplier/top	1695.887	1619.34	2746218.098	1781.895	1692.38	3015642.946	1726.972	1642.82	2837103.87	1798.997	1650.9	2969964.21	1636.313	1626.04	2660711.03
/benchmarks/random_control/mem_ctrl/mem_ctrl	1645.178	515.5	848089.2786	2005.523	624.04	1251526.411	1399.578	424.67	594358.763	1530.433	449.96	688633.831	1479.753	418.23	618877.247
/OPDB/picorv32	1520.329	344.63	523951.1477	1447.473	363.8	526590.7563	1528.815	341.08	521448.233	1478.951	362.14	535587.474	1455.886	346.75	504828.428
/vtr/RLE_BlobMerging	1519.105	806.13	1224595.931	1575.092	859.4	1353634.043	1439.221	783.26	1127284.2	1462.141	774.85	1132939.73	1474.359	783.71	1155469.7
/OPDB/dynamic_node_top_wrap	1374.967	400.55	550742.9834	1405.891	417.44	586875.1637	1398.834	381.79	534060.962	1389.212	401.73	558087.952	1363.726	378.35	515965.618
/quip/oc_vga_lcd	1304.735	492.54	642634.1868	1317.901	484.88	639023.7205	1312.448	474.83	623189.608	1316.341	496.62	653721.118	1293.392	462.76	598529.98
/iwls05/faraday/dma_top	1259.041	319.33	402049.6638	1293.144	333.87	431741.9659	1195.429	299.59	358138.507	1193.854	323.54	386259.567	1195.254	312.86	373947.109
/benchmarks/arithmetic/square/top	1232.054	667.26	822100.1759	1398.193	674.22	942689.5752	1230.523	655.79	806964.571	1383.496	662.29	916275.696	1215.403	663.3	806177.06
/quip/oc_aes_core	1223.481	378.76	463405.548	1336.024	386.81	516787.332	1174.346	361.26	424244.27	1133.522	359.47	407467.187	1152.753	364.53	420213.093
/quip/oc_mem_ctrl	1208.172	406.68	491339.2592	1136.744	422.89	480717.781	1111.346	398.25	442593.505	1118.68	424.05	474376.102	1099.522	386.17	424602.226
/iwls05/opencores/aes_cipher_top	1200.75	377.16	452875.0492	1265.296	393.31	497653.6221	1140.681	363.2	414295.294	1123.141	382.63	429747.492	1155.844	363.07	419652.308
/vtr/diffeq_paj_convert	1197.878	1637.46	1961477.662	1289.528	1678.27	2164176.313	1152.491	1521.14	1753099.67	1130.694	1566.63	1771378.48	1133.23	1551.99	1758762.4
/iwls05/opencores/usbf_top	1145.405	359.32	411566.8459	1137.371	443.38	504287.6431	1136.526	347.3	394715.327	1134.091	358.14	406163.244	1112.06	341.35	379601.79
/vtr/diffeq_f_systemC	1144.472	1422.12	1627576.06	1240.16	1453.87	1803031.73	1109.334	1363.86	1512976.1	1059.879	1406.98	1491227.88	1055.446	1383.07	1459755.97
/iwls05/opencores/ac97_top	1138.96	182.35	207689.433	1162.886	195.36	227181.4482	1149.779	185.99	213847.356	1133.07	187.77	212756.573	1127.573	186.43	210213.517
/quip/oc_wb_dma	1099.784	356.02	391545.0865	1111.506	360.9	401142.6175	1130.796	355.21	401669.914	1122.66	374.49	420424.937	1109.83	346.33	384367.279
/iwls05/iscas/s35932_bench	1076.208	169.44	182352.7027	1097.247	170.76	187365.9069	1021.446	163.52	167026.79	1038.096	188.73	195919.857	1010.788	167.8	169610.168
/iwls05/iscas/s38417_bench	977.6619	347.71	339942.8182	993.3354	354.3	351938.7312	971.6841	331.22	321841.206	971.5529	334.73	325207.895	962.3529	328.98	316594.856
/vtr/mkSMAdapter4B	887.5137	280.22	248699.1013	894.1039	305.31	272978.8629	857.2748	301	258039.722	865.1772	308.63	267019.634	838.4666	301.52	252814.456
/iwls05/iscas/s38584_bench	841.6159	219.47	184709.4455	841.2368	209.69	176398.9526	843.1906	211.99	178747.966	818.944	228.29	186956.73	822.4578	212.72	174953.223
/OPDB/gng	795.339	600.95	477958.9666	807.4987	583.32	471030.1481	784.9726	580.01	455291.964	746.3939	595	444104.389	738.652	564.24	416776.977
/quip/oc_ethernet	753.2174	326.97	246279.4861	771.3403	366.65	282811.9276	672.5171	362.85	244022.822	686.6451	333.46	228968.674	674.602	306.66	206873.455
/vtr/sha1	741.4659	836.1	619939.6281	781.1818	855.82	668551.0158	786.2411	826.3	649670.995	734.0884	850.7	624489.008	728.8688	849.38	619086.553
/iwls05/opencores/mc_top	687.1408	360.01	247377.5637	672.5171	392.63	264050.378	680.2153	377.51	256788.082	683.2042	379.57	259323.823	676.8473	366.03	247746.429
/benchmarks/random_control/arbiter/top	679.7779	191	129837.5825	428.5208	207.17	88776.64999	649.2328	196.13	127334.033	607.7673	199.84	121456.217	691.2378	187.77	129793.722
/iwls05/opencores/aes	651.4052	522.79	340548.1444	648.5913	548.32	355635.58	654.0734	500	327036.69	611.2373	496.36	303393.765	615.1156	468.53	288200.12
/quip/barrel64	618.8627	278.96	172637.9329	660.3428	263.79	174191.8217	393.3247	260.42	102429.608	505.3574	277.61	140292.262	268.5344	242.69	65170.623
/benchmarks/random_control/voter/top	612.535	867.08	531116.8088	619.4459	803.12	497489.3711	567.3953	827.93	469763.571	599.7775	811.7	486839.36	532.5199	805.33	428854.264
/quip/oc_minirisc	501.3333	382.46	191739.9309	491.8417	419.13	206145.6168	506.1739	386.36	195565.329	500.3564	370.22	185241.958	495.7929	385.32	191038.917
/benchmarks/arithmetic/sin/top	490.2233	1660.26	813898.2025	489.334	1625.4	795363.4186	479.6966	1658.69	795667.92	464.0814	1653.62	767414.283	440.2431	1615.52	711221.533
/iwls05/opencores/tv80s	467.5077	508.23	237601.4363	506.3197	489.44	247813.0924	430.9702	506.64	218346.75	440.5347	505.73	222791.612	436.3211	473.11	206427.864
/quip/mux8_128bit	453.1464	175.76	79645.00915	470.2925	173.01	81365.29989	445.6231	158.39	70582.2441	454.2107	166.12	75453.4861	453.1464	166.12	75276.678
/quip/fip_risc8	426.7566	524.26	223731.4114	415.1947	534.18	221788.6803	423.3011	503.78	213250.645	422.6159	493.23	208446.827	418.5189	509.51	213239.562
/quip/mux64_16bit	422.8054	208.73	88252.17323	432.3699	231.71	100184.427	428.7978	198.81	85249.2884	429.381	226.94	97443.7216	422.8054	208.73	88252.1732
/koios/reduction_layer	377.9865	247.58	93581.89594	372.7377	269.03	100277.6218	365.054	260.76	95191.4896	359.7469	256.7	92347.0326	371.0902	244.52	90738.9642
/quip/oc_des_des3area	369.9675	357.65	132318.876	313.5137	384.91	120674.5733	309.8542	330.27	102335.533	303.7014	351.07	106620.45	295.0409	347.7	102585.714
/iwls05/opencores/wb_dma_top	345.2981	246.2	85012.40108	354.746	220.41	78189.56057	351.5821	219.89	77309.3915	346.3479	215.54	74651.8255	345.1086	219.74	75834.1629
/OPDB/dynamic_node_top_wrap_para	314.9717	302.58	95304.14758	320.4101	316.86	101525.1364	327.5689	323.93	106109.379	323.2823	313.93	101488.023	321.2266	291.05	93492.9888
/OPDB/sparc_ifu_esl	271.9024	614.69	167135.6979	293.4808	640.66	188021.4215	276.7284	633.8	175390.459	297.3883	631.57	187821.503	270.4736	640.85	173332.993
/benchmarks/arithmetic/bar/top	250.7177	164.93	41350.86663	257.891	176.79	45592.55661	257.4682	169.95	43756.7235	241.4885	168.53	40698.0636	233.4695	165.46	38629.8701
/iwls05/opencores/systemcdes/des	237.931	393.39	93599.68396	246.6207	393.39	97018.11717	244.2004	406.84	99350.4989	242.5383	385.73	93554.2985	233.3966	376.37	87843.4934
/quip/oc_des_area_opt	231.268	306.63	70913.69457	263.1107	270.26	71108.29238	235.1608	300.61	70691.6941	246.6936	286.06	70569.1712	240.2055	268.36	64461.5477
/quip/mux8_64bit	229.8245	166.91	38360.01297	244.8857	182.68	44735.71493	224.4008	164.8	36981.2476	229.8245	166.91	38360.013	229.8245	166.91	38360.013
/iwls05/opencores/spi_top	226.6024	420.72	95336.1449	242.3342	421.39	102117.1997	217.3003	394.29	85679.3428	210.5789	407.88	85890.9376	217.971	385.52	84032.1795
/quip/mux32_16bit	219.3707	187.45	41121.03303	221.4994	218.49	48395.39386	224.5903	202.1	45389.7025	219.5165	175.81	38593.1915	219.3707	185.05	40594.5434
/benchmarks/arithmetic/max/top	200.0084	475.64	95132.01488	269.6717	499.71	134757.6352	196.2322	474.47	93106.3014	206.3653	503.86	103979.231	189.4379	484.03	91693.6461
/quip/fip_cordic_rca	199.017	475.82	94696.26846	187.3676	457.61	85741.27783	144.619	484.13	70014.4057	178.5467	461.71	82436.7872	160.0738	451.91	72338.9595
/quip/barrel32	193.637	232.73	45065.13436	120.3287	191.38	23028.51426	121.0577	206.27	24970.58	121.8013	204.72	24935.1662	113.0387	204.11	23072.3372
/quip/fip_cordic_cla	157.3474	455.6	71687.45676	166.7515	449.44	74944.77618	153.5857	404.7	62156.1405	159.8405	420	67133.0264	155.0729	392.92	60931.2356
/quip/oc_video_compression_systems_huffman_dec	137.8247	311.98	42998.56239	137.7518	300.93	41453.66121	128.0124	286.6	36688.3538	130.9576	284.09	37203.7332	123.1427	281.42	34654.813
/iwls05/iscas/s13207_bench	137.6935	156.11	21495.33525	139.6181	154.08	21512.35377	133.7278	180.7	24164.6061	136.9499	159.48	21840.7764	135.5503	157.23	21312.5672
/benchmarks/arithmetic/adder/top	129.5579	260.59	33761.48795	135.4774	254.11	34426.1522	155.3207	263.97	41000.0157	188.0383	240.38	45200.6369	148.9639	252.85	37665.512
/iwls05/iscas/s5378_bench	122.8802	237.89	29231.98006	128.858	244.47	31501.92504	121.5097	236.22	28703.0261	123.5509	228.81	28269.686	123.201	229.83	28315.2858
/vtr/memset	111.3766	246.91	27500.00075	112.0327	239.64	26847.52054	113.7532	245.05	27875.2114	114.9196	238.59	27418.6573	108.6502	244.63	26579.0882
/quip/oc_video_compression_systems_huffman_enc	110.2977	209.02	23054.42525	110.3123	197.72	21810.9438	109.7582	204.06	22397.2665	114.0156	192.1	21902.3968	109.1896	186.22	20333.291
/quip/oc_rtc	110.2977	249.43	27511.55506	105.0489	268.63	28219.28574	116.9753	236.8	27699.7605	109.4521	229.95	25168.5012	103.6638	225.66	23392.7729
/benchmarks/random_control/dec/dec	107.5275	76.52	8228.0043	107.5275	76.52	8228.0043	107.5275	76.52	8228.0043	107.5275	76.52	8228.0043	107.5275	76.52	8228.0043
/benchmarks/random_control/i2c/i2c	103.8388	112.48	11679.78372	120.0663	112.85	13549.48196	103.9117	115.68	12020.5008	124.2945	106.32	13214.9911	102.8036	111.62	11474.9356
/quip/os_sdram16	100.3833	85.4	8572.73382	100.5437	85.4	8586.430272	100.9957	85.4	8625.02936	99.8001	85.4	8522.92854	100.2812	85.4	8564.0179
/iwls05/iscas/s9234_1_bench	99.49392	231.14	22997.02467	102.235	230.72	23587.64997	104.5678	225.6	23590.4867	100.4999	230.03	23118.0012	98.15256	233.33	22901.9368
/quip/oc_i2c	94.17222	221.69	20877.03945	98.83782	229.67	22700.08212	92.75796	230.21	21353.81	92.3643	234.14	21626.1772	90.9063	223.94	20357.5568
/iwls05/opencores/i2c_master_top	91.81026	233.92	21476.25602	98.07966	225.86	22152.27201	91.59156	225.53	20656.6445	92.61216	223.47	20696.0394	92.40804	224.56	20751.1495
/quip/oc_sdram	90.13356	183.33	16524.18555	87.08634	182.91	15928.96245	86.50314	194.88	16857.7319	86.08032	189.23	16288.979	86.06574	178.82	15390.2756
/iwls05/opencores/simple_spi_top	79.8984	186.75	14921.02601	83.51424	197.93	16529.97333	81.75006	171.91	14053.6526	83.835	187.52	15720.739	80.7003	171.47	13837.6803
/quip/oc_gpio	74.53296	137.16	10222.94079	75.91806	136.08	10330.9296	70.61094	135.4	9560.72128	68.0886	171.76	11694.8979	72.4626	134.88	9773.75549
/iwls05/iscas/s15850_bench	74.48922	188.41	14034.51394	76.38462	199.31	15224.21861	72.41886	194.84	14110.0907	73.14786	190.33	13922.2322	72.79794	191.98	13975.7485
/OPDB/fpu	73.7019	172.8	12735.68815	47.41416	159.1	7543.592697	25.32546	124.54	3154.03266	14.7258	125.76	1851.91661	26.01072	118.51	3082.53043
/iwls05/opencores/sasc_top	70.11522	169.53	11886.63308	71.55864	165.44	11838.6614	66.1203	168.37	11132.6747	63.91872	164.24	10498.0104	63.26262	159.36	10081.531
/vtr/sv_chip3_hierarchy_no_mem	67.66578	195.9	13255.72611	66.47022	188.06	12500.38939	63.84582	191.8	12245.6281	63.77292	193.92	12366.8445	61.87752	186.42	11535.2071
/benchmarks/random_control/cavlc/top	67.18464	118.11	7935.17783	61.58592	123.2	7587.385344	60.52158	125.17	7575.48617	56.55582	122.77	6943.35802	56.6433	119.98	6796.06313
/iwls05/iscas/s1423_bench	65.95992	283.37	18691.06253	69.5466	274.55	19094.01903	71.23788	295.88	21077.8639	69.82362	260.06	18158.3306	65.93076	259.96	17139.3604
/OPDB/sparc_ifu_esl_fsm	61.58592	304.88	18776.31498	59.67594	283.34	16908.58056	56.67246	266.5	15103.2103	53.24616	301.05	16029.7562	51.03	240.13	12253.8337
/benchmarks/random_control/priority/top	53.26074	134.89	7184.341219	49.7178	188.07	9350.426646	49.36788	140.88	6954.94693	53.42112	137.14	7326.1724	43.36092	129.34	5608.30139
/iwls05/opencores/usb_phy	53.12952	170.74	9071.334074	52.37136	165.83	8684.742463	53.37738	151.46	8084.53782	53.2899	166.86	8891.95255	53.23158	157.88	8404.20169
/quip/barrel16	51.87564	215.26	11166.75027	43.85664	172.67	7572.726029	39.59928	174.84	6923.53812	38.81196	174.63	6777.7324	33.95682	169.88	5768.58458
/quip/xbar_16x16	43.8858	104.47	4584.749526	43.8858	104.47	4584.749526	48.08484	116.74	5613.42422	43.8858	104.47	4584.74953	42.69024	104.47	4459.84937
/iwls05/iscas/s1494_bench	43.49214	188.52	8199.138233	41.8446	185.9	7778.91114	40.37202	188.42	7606.89601	37.7622	210.7	7956.49554	36.55206	186.65	6822.442
/iwls05/opencores/pcm_slv_top	41.6988	146.94	6127.221525	41.05728	138.16	5672.473667	43.1568	156.93	6772.59647	43.0839	156.93	6761.15627	41.59674	146.94	6112.22483
/iwls05/iscas/s1488_bench	40.21164	186	7479.36504	39.10356	186.91	7308.8464	40.37202	214.92	8676.75454	39.70134	196.88	7816.39982	37.96632	182.57	6931.51104
/iwls05/iscas/s1196_bench	37.73304	194.92	7354.924157	37.22274	196.92	7329.901961	40.32828	186.15	7507.10932	36.56664	184.54	6748.00775	34.71498	179.02	6214.67572
/iwls05/iscas/s1238_bench	37.09152	199.44	7397.532749	43.78374	180.31	7894.646159	38.43288	197.07	7573.96766	35.15238	207.15	7281.81552	37.31022	186.03	6940.82023
/quip/barrel16a	33.91308	176.32	5979.554266	32.06142	177.48	5690.260822	34.81704	166.47	5795.99265	32.76126	165.6	5425.26466	30.95334	166.42	5151.25484
/iwls05/iscas/s838_1_bench	31.69692	174.8	5540.621616	31.94478	175.52	5606.947786	32.49882	210.38	6837.10175	31.65318	179.44	5679.84662	31.52196	199.61	6292.09844
/quip/oc_fcmp	26.95842	160.7	4332.218094	25.82118	164.82	4255.846888	32.2947	167.58	5411.94583	27.45414	167.9	4609.55011	26.85636	159.87	4293.52627
/iwls05/iscas/s820_bench	23.37174	172.7	4036.299498	21.41802	169.11	3622.001362	20.35368	172.06	3502.05418	20.67444	171.61	3547.94065	19.77048	172.54	3411.19862
/benchmarks/random_control/ctrl/top	22.84686	65.92	1506.065011	17.94798	83.14	1492.195057	20.23704	67.26	1361.14331	20.00376	79.26	1585.49802	17.7876	63.96	1137.6949
/iwls05/iscas/s832_bench	21.0681	169.09	3562.405029	20.9952	170.5	3579.6816	20.71818	169.01	3501.5796	18.86652	176.46	3329.18612	18.8811	166.77	3148.80105
/benchmarks/random_control/int2float/top	20.42658	99.62	2034.8959	18.99774	103.36	1963.606406	18.5895	100.87	1875.12287	20.89314	106.3	2220.94078	16.94196	97.34	1649.13039
/iwls05/iscas/s713_bench	19.63926	146.15	2870.277849	19.26018	141.32	2721.848638	17.61264	148.98	2623.93111	15.5277	152.37	2365.95565	16.63578	145.32	2417.51155
/iwls05/iscas/s526_bench	19.4643	148.61	2892.589623	18.38538	153.03	2813.514701	18.954	155.35	2944.5039	17.87508	151.2	2702.7121	17.37936	148.35	2578.22806
/iwls05/iscas/s526n_bench	18.5166	154.54	2861.555364	19.27476	157.71	3039.8224	18.06462	143.59	2593.89879	18.06462	148.03	2674.1057	17.30646	143.49	2483.30395
/iwls05/iscas/s382_bench	18.4437	156.06	2878.323822	18.5895	150.87	2804.597865	16.73784	150.6	2520.7187	17.24814	149.15	2572.56008	15.99426	147.86	2364.91128
/iwls05/iscas/s641_bench	18.26874	156.02	2850.288815	20.95146	182.46	3822.803392	18.74988	176.81	3315.16628	19.33308	145	2803.2966	16.75242	148.78	2492.42505
/iwls05/iscas/s444_bench	17.9334	152.13	2728.208142	16.75242	153.2	2566.470744	17.01486	152.94	2602.25269	17.33562	151.9	2633.28068	16.44624	145.76	2397.20394
/iwls05/iscas/s510_bench	17.23356	166.02	2861.115631	16.9857	160.85	2732.149845	16.73784	163.31	2733.45665	17.43768	166.55	2904.2456	16.44624	165.5	2721.85272
/iwls05/iscas/s400_bench	16.73784	149.23	2497.787863	17.75844	148.42	2635.707665	17.55432	153.8	2699.85442	19.44972	154.01	2995.45138	16.9857	144.59	2455.96236
/iwls05/iscas/s420_1_bench	16.63578	146.99	2445.293302	15.27984	184.43	2818.060891	15.46938	179.66	2779.22881	16.37334	157.58	2580.11092	14.8716	148.25	2204.7147
/OPDB/sparc_ifu_esl_counter	15.57144	189.6	2952.345024	12.58254	203.02	2554.507271	14.75496	208.77	3080.393	13.6323	196.49	2678.61063	14.18634	195.84	2778.25283
/OPDB/sparc_ifu_esl_htsm	15.42564	115.13	1775.953933	10.68714	116.65	1246.654881	12.8304	114.84	1473.44314	9.9873	120.39	1202.37105	10.62882	77.5	823.73355
/iwls05/iscas/s344_bench	13.90932	159.39	2217.006515	15.06114	162.78	2451.652369	12.93246	164.58	2128.42427	13.80726	159.59	2203.50062	13.25322	155.4	2059.55039
/iwls05/iscas/s349_bench	13.79268	158.8	2190.277584	15.80472	155.55	2458.424196	14.4342	148	2136.2616	13.7781	153.82	2119.34734	12.8304	148.11	1900.31054
/benchmarks/random_control/router/top	11.38698	108.79	1238.789554	10.73088	121.62	1305.089626	9.6957	116.11	1125.76773	9.44784	126.22	1192.50636	7.01298	108.85	763.362873
/iwls05/iscas/s298_bench	11.09538	146.99	1630.909906	11.70774	153.18	1793.391613	10.55592	144.19	1522.0581	10.99332	152.62	1677.8005	10.7163	140.17	1502.10377
/iwls05/iscas/s386_bench	10.67256	170.02	1814.548651	12.05766	144.06	1737.0265	10.46844	171.34	1793.66251	11.43072	164.52	1880.58205	10.04562	139.1	1397.34574
/OPDB/sparc_ifu_esl_lfsr	10.3518	117.2	1213.23096	11.99934	108.6	1303.128324	11.8827	118.26	1405.2481	11.8827	118.26	1405.2481	11.8827	118.26	1405.2481
/OPDB/sparc_ifu_esl_stsm	8.51472	91.91	782.5879152	12.8304	83.66	1073.391264	10.206	85.83	875.98098	7.75656	86.38	670.011653	7.30458	86.38	630.96962
/OPDB/sparc_ifu_esl_shiftreg	8.04816	73.36	590.4130176	8.04816	73.36	590.4130176	8.04816	73.36	590.413018	8.04816	73.36	590.413018	8.04816	73.36	590.413018
/iwls05/iscas/s208_1_bench	7.8732	158.08	1244.595456	7.63992	162.01	1237.743439	7.75656	164.14	1273.16176	8.06274	155.69	1255.28799	6.95466	133.02	925.108873
/quip/ts_mike_fsm	6.4881	126.27	819.252387	7.12962	132.49	944.6033538	7.4358	125.3	931.70574	7.4358	125.3	931.70574	7.04214	116.21	818.367089
/OPDB/sparc_ifu_esl_rtsm	4.6656	67.2	313.52832	5.84658	73.44	429.3728352	5.65704	65.92	372.912077	5.65704	65.92	372.912077	5.65704	65.92	372.912077
/iwls05/iscas/s27_bench	2.09952	100.2	210.371904	2.09952	100.2	210.371904	2.09952	100.2	210.371904	2.09952	100.2	210.371904	2.09952	100.2	210.371904
/OPDB/sparc_mul_top_nospu_wrap	1.19556	71.08	84.9804048	1.19556	71.08	84.9804048	1.19556	71.08	84.9804048	1.19556	71.08	84.9804048	1.19556	71.08	84.9804048