C clang versus Chapel fastest performance

How the programs are written matters! Always look at the source code.

If the fastest programs are hand-written vector instructions, does the host language matter? You might be more interested in the less optimised programs — more seconds, less gz source code.

fannkuch-redux
source	secs	mem	gz	cpu secs
* C clang #6	2.24	19,772	1582	8.86
* C clang #4	13.37	19,772	1190	13.37
C clang #5	8.09	19,772	917	31.94
* C clang #2	8.67	19,772	1563	34.02
C clang #3	34.16	19,772	574	34.15
C clang #8	36.39	19,492	483	36.38
C clang	36.51	19,772	514	36.50
Chapel #2	9.34	19,436	737	37.22
n-body
source	secs	mem	gz	cpu secs
* C clang #9	2.20	19,772	1639	2.20
Chapel #3	3.86	19,436	983	3.87
Chapel #4	4.34	19,436	1027	4.34
* C clang #5	4.86	19,772	1436	4.86
C clang #6	5.16	19,772	1186	5.16
C clang	5.37	19,772	1179	5.37
C clang #2	5.42	19,772	1270	5.42
* C clang #4	5.60	19,772	1398	5.59
Chapel #2	5.67	19,436	977	5.67
C clang #8	5.72	19,492	1193	5.72
C clang #3	6.12	19,772	1214	6.12
C clang #7	6.77	19,772	1250	6.77
spectral-norm
source	secs	mem	gz	cpu secs
* C clang #6	0.38	19,772	1203	1.52
* C clang #4	0.72	19,772	1145	2.86
* C clang #2	0.72	19,772	941	2.87
* C clang #5	0.72	19,772	576	2.88
Chapel #2	0.73	19,436	348	2.88
C clang #8	5.40	19,772	389	5.40
C clang #3	1.46	19,772	470	5.77
Chapel	1.46	19,436	322	5.79
* C clang #7	1.82	19,772	906	7.24
mandelbrot
source	secs	mem	gz	cpu secs
C clang #9	1.22	34,332	700	4.85
Chapel #3	1.29	37,760	582	5.12
C clang #5	1.53	34,864	788	6.09
* C clang #4	3.47	34,328	805	13.82
* C clang	3.49	33,172	828	13.93
* C clang #7	3.53	34,216	1000	14.00
* C clang #3	3.55	33,108	769	14.15
C clang #8	25.27	19,496	412	25.27
Chapel	6.63	37,632	447	26.46
* C clang #6		Make Error
fasta
source	secs	mem	gz	cpu secs
C clang #9	0.78	19,772	1469	0.78
Chapel #6	0.79	19,436	1108	0.80
C clang #5	1.19	19,772	1286	1.19
C clang #7	2.43	19,772	1299	2.43
C clang #6	1.03	19,772	1936	3.03
C clang #2	0.83	19,772	2276	3.28
C clang	3.46	19,772	1213	3.46
Chapel #5	1.03	19,436	1391	3.78
Chapel #3	4.02	19,436	1219	4.03
C clang #8	4.60	19,772	825	4.60
C clang #4		Make Error
k-nucleotide
source	secs	mem	gz	cpu secs
C clang	6.30	129,228	1512	13.92
Chapel #4	4.97	191,872	1071	17.19
Chapel #3	5.89	192,000	1040	20.88
reverse-complement
source	secs	mem	gz	cpu secs
* C clang #7	0.45	499,996	1899	0.63
C clang #9	0.46	499,976	1551	0.70
C clang #2	1.16	994,536	757	1.69
C clang #5	2.08	994,108	654	2.07
Chapel #9	0.92	1,057,024	1260	2.65
Chapel #5	1.83	2,133,376	554	3.18
Chapel #3	1.81	2,121,216	709	3.18
* C clang #4	3.74	500,452	716	3.74
C clang #6	1.03	703,416	828	4.07
binary-trees
source	secs	mem	gz	cpu secs
C clang #2	1.66	170,236	816	5.36
C clang #3	1.89	134,272	842	5.88
Chapel #5	2.04	1,775,616	646	7.28
C clang	17.79	264,064	660	17.79
C clang #8	19.63	285,536	607	19.62
Chapel #3	8.47	300,672	494	25.56
Chapel #4	7.11	477,824	489	26.09
C clang #5	9.25	351,328	914	33.65
pidigits
source	secs	mem	gz	cpu secs
C clang #2	0.74	19,772	422	0.74
C clang	0.74	19,772	459	0.74
* Chapel #4	0.75	19,436	515	0.76
Chapel #2	0.76	19,436	423	0.76
regex-redux
source	secs	mem	gz	cpu secs
C clang #5	0.82	154,568	1403	2.82
C clang #2	3.22	103,364	1121	3.22
C clang #4	0.98	154,488	1237	3.82
Chapel #3	3.03	404,608	503	3.96
C clang #3	1.75	105,848	1273	4.31
Chapel #2	3.53	404,608	498	4.47
C clang	Ubuntu clang version 18.1.8 (11)
Chapel	chpl version 2.2.0 built with LLVM version 18.1.3 Copyright 2020-2024 Hewlett Packard Enterprise Development LP Copyright 2004-2019 Cray Inc.